Skip to main content
archive Search
Submit Donate Log in

Press Enter to search · Advanced search

Optimization and Control

  • New submissions
  • Cross-lists
  • Replacements

See recent articles

Showing new listings for Monday, 16 November 2026

Total of 67 entries
Showing up to 2000 entries per page: fewer | more | all

New submissions (showing 28 of 28 entries)

[1] arXiv:2611.17447 [pdf, html, other]
Title: Distributionally Robust Risk Evaluation with Low-Rank Perturbations
Linglingzhi Zhu, Dongze Wu, Yao Xie
Subjects: Optimization and Control (math.OC); Machine Learning (stat.ML)

We study distributionally robust risk under perturbations restricted to low-rank matrices whose row and column spaces are shared across observations, focusing on how much worst-case risk this restriction sacrifices and how to compute the restricted worst case. Under moment or Wasserstein budgets, we derive quantitative bounds on the gap between the unrestricted and low-rank worst-case risks, characterize when the gap vanishes, and identify the smallest lossless row and column dimensions. To generate worst-case perturbations, we apply projected gradient ascent to a curvature-corrected concave problem and derive computable risk certificates. We establish global sublinear convergence for the corrected objective and linear convergence to a worst-case distribution when the budget constraint is sufficiently active.

[2] arXiv:2611.17477 [pdf, html, other]
Title: First-order characterizations of strong quasiconvexity with the exact modulus
John Matos Ascona, Jackeline C. Huaccha Neyra, Lizet Santa Cruz Calderón, Heraclio López-Lázaro
Comments: 21 pages, 1 figure, 1 table
Subjects: Optimization and Control (math.OC)

In this paper, we study first-order characterizations of strong quasiconvexity for functions defined on convex subsets of real normed spaces. For functions that are continuous along segments, an inequality involving the lower Dini derivative is a known necessary condition for strong quasiconvexity with a given modulus, whereas the established converse guarantees only half of that modulus. By means of a one-dimensional lemma, we prove that this inequality, and even its weaker version with the upper Dini derivative, ensures strong quasiconvexity with the same modulus, without requiring the domain to be open; this answers a question raised in the recent literature. The main result leads to a characterization with the same modulus in terms of the Clarke subdifferential for locally Lipschitz functions on Banach spaces, which answers another open question, and to a formula for the exact modulus of radial functions defined on Hilbert spaces. We also present some consequences in optimization, such as an optimality condition without Clarke regularity, an error bound with optimal constant and, for the projected subgradient method, the same convergence rate as in the strongly convex case, together with the exponential convergence of the associated flow. Finally, we show that the converse of a relation between proximal points and strong subdifferentials fails in general, but holds for a truncation of the function.

[3] arXiv:2611.17488 [pdf, html, other]
Title: Convergence Properties of PF-Exec: Homogeneous Structure, Anchored Closed Form, and Gain Dynamics
Hanzhang Jia, Liheng Zeng, Yiran Li, Yi Gao, Bo Ma
Subjects: Optimization and Control (math.OC); Machine Learning (cs.LG); Signal Processing (eess.SP)

This paper establishes a layered mathematical char acterisation of PF-Exec, the algorithm of the companion system paper. Demand-proportional bandwidth allocation yields an exact first-order homogeneous decomposition, from which the equal-load bandwidth target and its feasibility bound follow in closed form. The logarithmic knee view preserves strict load order and shares its global optimum with the Jain fairness index, while the demand-weighted centroid minimises the position potential. The damped softmax association block possesses a fixed point on the product of simplices. The truncated one-dimensional gain iteration admits a complete characterisation, with geomet ric convergence below the critical gain, initial-value-dependent bounded oscillation at it, and a family of bounded two-state cycles above it. A global invariant error ball and an explicit tracking recursion complete the execution guarantee. A numerical audit confirms the structural identities on every measured state and quantifies the local amplification of the coupled association and resource dynamics. Together, the results explain how closed form recovery directions and bounded adaptive gains provide stable execution without requiring a global Euclidean contraction assumption.

[4] arXiv:2611.17613 [pdf, other]
Title: Innovative Integrated Congestion Management Market: Synergizing Operational Flexibility of Transmission System Operators with Aggregator-Driven Resources
Arash Beiranvand
Subjects: Optimization and Control (math.OC)

This paper introduces an innovative congestion management market that seeks to integrate and harness the operational flexibility available to Transmission System Operators (TSO) in conjunction with the flexibility offered by aggregators. In the context of operational flexibility, this study investigates Network Topology Optimization (NTO) as a pivotal strategy for to address the transmission network congestion, specifically through the strategic switching of certain buses. To achieve this, an N-1 secured NTO is formulated using a Mixed Integer Linear Programming (MILP) methodology. Concurrently, acknowledging the crucial and growing importance of aggregator-provided flexibility in congestion management, the paper presents a novel integrated congestion management market. In this framework, aggregators submit bids, and the TSO, taking into account these bids alongside the potential for the NTO, clears the market with the goal of minimizing the overall cost of congestion management. The proposed integrated market is constructed using a linear optimization model, ensuring its practical applicability in real world scenarios. The effectiveness and efficiency of the proposed market design are assessed using the standard IEEE 39-bus system, with results indicating promising potential and applicability

[5] arXiv:2611.17656 [pdf, html, other]
Title: Optimal Dividend Problem for a Threshold Dependent Surplus Process with Bounded Dividend Rate
Zhongqin Gao, Panyu Wu, Xiaowen Zhou
Comments: 25 pages, 6 figures
Subjects: Optimization and Control (math.OC); Probability (math.PR)

We study optimal bounded-rate dividend payments for a diffusion surplus process whose drift and volatility change at a fixed surplus threshold. Optimal strategies admit an exhaustive classification into zero-threshold, positive single-threshold, two-boundary, and three-boundary configurations. The latter two configurations have separated dividend regions, which arise because the interface can reverse the monotonicity of the derivative of the value function with respect to the initial surplus. Multiband optimality requires an upward change in drift. Without prescribing a strategy in advance, we construct the unique bounded classical solution of the Hamilton-Jacobi-Bellman transmission problem by varying the derivative at zero surplus and selecting the unique value for which the resulting solution agrees in both value and derivative with a bounded solution at a fixed sufficiently large surplus level. Verification identifies this solution with the value function and determines the optimal feedback strategy. For each configuration, we derive necessary and sufficient optimality conditions and characterize the switching boundaries through explicit formulas or equations in one unknown obtained by matching the value function and its derivative. The analysis also shows how changes in the drift and volatility of the upper regime affect the optimal value and dividend strategy. Numerical analysis illustrates how model parameters affect the optimal dividend regions.

[6] arXiv:2611.17682 [pdf, html, other]
Title: Agent-written heuristics versus commercial solvers on unsolved job-shop benchmark instances
Andre Albrecht, Enzo Morosini Frazzon, Leandro C. Coelho, Bruno Vieira
Comments: 27 pages, 4 figures, 14 tables. Submitted to International Transactions in Operational Research. Data, schedules, checkers and code: this https URL
Subjects: Optimization and Control (math.OC); Artificial Intelligence (cs.AI)

Commercial solvers are the standard tool for difficult scheduling problems. We study whether a frontier large language model, used as a coding agent with open-source tools only, can match them in quality and speed. We report every run, and independent checkers verify every schedule. On the 100,000-operation job shop tai_100_1000_1, an agent-written solver reached 525,515, below OptalCP, Hexaly, and CP Optimizer. Unchanged, it improved all ten instances of the family and closed two. With equal budgets on the same workstation, it ended 0.0-1.8% above the lower bound, against 2.6-5.8% for CP Optimizer and 2.4-4.4% for Hexaly. On the Behnke-Geiger flexible job shops, an agent-written tabu search improved 11 of 32 open instances. With two transport robots, an agent-written solver improved seven of ten MKT instances. New lower bounds show that four published MKT values cannot be attained on the distributed instance files. Agent-run improvement campaigns closed eight further instances.

[7] arXiv:2611.17714 [pdf, html, other]
Title: Online Carbon-Aware Power Flexibility Quantification of Distributed Data Centers
Xiaowei Wang, Yue Chen
Comments: 12 pages, 12 figures
Subjects: Optimization and Control (math.OC)

Geographically distributed data centers (DCs) can provide spatio-temporal power flexibility to electric grids through the coordinated management of computing resources and energy. Beyond their energy consumption, the carbon footprints of DCs are non-negligible, which constrain the power flexibility that DCs can offer to the power grid. To address this challenge, this paper proposes a carbon-aware power flexibility quantification framework for DCs. We first develop an optimization model that characterizes DC power flexibility in a grid-usable form, namely the Cartesian product of adjustable power demand intervals across time slots. An online algorithm based on Lyapunov optimization is then designed to solve the flexibility quantification model in real time under uncertain workload arrivals and fluctuating grid carbon intensity. Furthermore, an integrated disaggregation and feedback mechanism is devised to translate the power system operator's dispatch commands into real-time DC operating decisions and to capture their impact on future DC flexibility. Several desirable properties of the proposed model and method are proven, including disaggregation feasibility, bounded workload processing delays, and a bounded suboptimality gap. Numerical simulations demonstrate that our approach effectively unlocks DC power flexibility while satisfying carbon emission constraints, outperforming baseline algorithms.

[8] arXiv:2611.17727 [pdf, html, other]
Title: Proximal Adam algorithm for nonsmooth composite optimization and linear convergence under proximal PL geometry
Xiaopeng Zhu, Zhongyi Huang
Comments: 27 pages,3figures
Subjects: Optimization and Control (math.OC); Numerical Analysis (math.NA)

We analyze Adam-type methods for nonsmooth composite optimization problems $\min_x\Phi(x)=f(x)+h(x)$, where $f$ is smooth and $h$ is convex, proper, and lower semicontinuous. The nonsmooth term is treated by a metric-weighted proximal map, while the smooth term is preconditioned by the Adam second-moment matrix. We prove a composite descent lemma whose constants depend explicitly on the spectral bounds of the adaptive metric and a residual--step equivalence that converts proximal step decay into subgradient stationarity. These yield an $O(1/N)$ bound on the minimum proximal stationarity residual for the deterministic proximal normalized gradient method without convexity of $f$, and the same $O(1/N)$ rate for proximal Adam under quantitative momentum-alignment and residual-coherence conditions together with a metric-compatibility condition. By estimating the momentum error directly, the same $O(1/N)$ rate is also obtained without the momentum-alignment and residual-coherence assumptions. When the composite objective satisfies a proximal Polyak--Łojasiewicz inequality, both algorithms and the relaxed proximal Adam scheme converge linearly with explicit contraction factors. Numerical experiments on a diagonal Lasso problem, two-dimensional image TV denoising, and two-dimensional elastic parameter identification illustrate the predicted stationarity and linear rates.

[9] arXiv:2611.17736 [pdf, html, other]
Title: High-probability convergence of adaptive, proximal, and clipped Adam-type methods for nonconvex stochastic optimization
Xiaopeng Zhu, Zhongyi Huang
Comments: 30 pages, 7 figures
Subjects: Optimization and Control (math.OC); Numerical Analysis (math.NA)

This paper develops a unified high-probability analysis of adaptive, proximal, and clipped first-order methods for nonconvex stochastic optimization. For a smooth objective whose stochastic gradient error has sub-Gaussian norm, the normalized adaptive gradient method attains the standard stochastic nonconvex complexity of the square root of the number of iterations, with an additional logarithmic dependence on the inverse confidence level that is inversely proportional to the number of iterations. The proof uses an adaptive descent decomposition together with Freedman and Bernstein concentration inequalities. The same guarantee is established for the full adaptive moment method under an explicit momentum-alignment condition. The analysis is extended to projected and proximal updates through a stationarity measure based on the subdifferential distance, to objectives satisfying the Polyak-Lojasiewicz condition, and to heavy-tailed noise through truncation. The framework is applied to stochastic inverse problems in elasticity, and numerical experiments illustrate the theoretical scaling and the practical behavior of the methods.

[10] arXiv:2611.17745 [pdf, html, other]
Title: The Polyak-Łojasiewicz Paper: A 10-Year Retrospective
Hamed Karimi, Julie Nutini, Mark Schmidt
Comments: 20 pages, 6 figures
Subjects: Optimization and Control (math.OC); Machine Learning (cs.LG); Machine Learning (stat.ML)

In 2016, we wrote a paper introducing the phrase "Polyak-Lojasiewicz (PL) condition" for an old idea, unsure of how readers would react. The paper ultimately had a larger impact than we expected, and we are surprised and honoured to receive the 2026 PKDD ECML Test of Time Award. With 10 years of hindsight, this note offers an informal and candid retrospective on the PL condition. We take a look back at the origins of the paper, give an AI-assisted refined version of the main theorem, review criticisms of the work, and discuss a variety of related ideas and extensions.

[11] arXiv:2611.17813 [pdf, html, other]
Title: An accelerated multiobjective gradient flow with implicit Hessian-driven damping
Wen Jiang, Hao Luo
Subjects: Optimization and Control (math.OC); Dynamical Systems (math.DS)

In this paper, we propose an accelerated multiobjective gradient flow with implicit Hessian-driven damping for unconstrained convex multiobjective optimization problems in a real Hilbert space. The continuous flow is formulated as a system of two first-order equations involving only a single multiobjective projection operator and suitable time-scaling factors. Formally, it can also be expressed as a second-order system with explicit multiobjective Hessian-driven damping, which corresponds to the usual Hessian-driven damping introduced by Alvarez et al. [J. Math. Pures Appl., 81(8):747--779, 2002]. The first-order formulation, however, involves no Hessian term and can be regarded as an extension of the Hessian-driven Nesterov accelerated gradient flow by Chen and Luo [arXiv:1912.09276, 2019]. Assuming that the objective gradients are Lipschitz continuous on bounded sets and that the common lower level sets are bounded, we establish global existence in finite dimensions under suitable time-scaling conditions. An a priori energy estimate confines the solutions to a fixed bounded set, and a Lyapunov argument yields fast decay of a merit function and the KKT residual. Under an additional bound on the time-scaling factors, the trajectory of every global solution converges weakly to a weak Pareto point. Numerical tests illustrate the rapid residual decay and reduced oscillations.

[12] arXiv:2611.17949 [pdf, html, other]
Title: Auto-conditioned Gradient Descent Methods for Multiobjective Optimization
Shuaichenji Jiang
Subjects: Optimization and Control (math.OC)

This paper proposes two linesearch-free algorithms based on steepest descent methods for smooth multiobjective optimization: auto-conditioned steepest descent method for multiobjective optimization (AC-SDM) and auto-conditioned normalized gradient descent method for multiobjective optimization (AC-NGDM). Both algorithms estimate the local Lipschitz constant from the function values and the first-order information obtained in previous iterations. AC-SDM uses a shared estimate of the Lipschitz constant to construct a scaled steepest descent direction. For AC-NGDM, we use each function's local Lipschitz estimate to normalize the gradient of the corresponding function and construct the descent direction. We prove that every accumulation point of any sequence generated by both algorithms is a Pareto critical point. We also establish convergence rates of $O(1/\sqrt{k})$ for nonconvex objectives and $O(1/k)$ for convex objectives. Numerical results on several test problems demonstrate that AC-SDM and AC-NGDM require fewer iterations and less CPU time than the multiobjective steepest descent method with an Armijo line search and a max-type nonmonotone line search.

[13] arXiv:2611.17987 [pdf, html, other]
Title: The Train Unit Shunting Problem with Service Scheduling and Workforce Assignment: classical and quantum optimization approaches
Alessia Ciacco, Francesca Guerriero, Eneko Osaba
Comments: Train Unit Shunting Problem, Railway Optimization, Mixed-Integer Linear Programming, Quantum Annealing, Hybrid Quantum-Classical Optimization
Subjects: Optimization and Control (math.OC); Emerging Technologies (cs.ET); Quantum Physics (quant-ph)

Efficient railway rolling stock management requires the coordinated planning of train-unit assignment, shunting movements, track occupation, service operations, and workforce allocation. These interdependent decisions give rise to a challenging combinatorial optimization problem, particularly as the size of the fleet and the planning horizon increase. This paper introduces the Train Unit Shunting Problem with Service Scheduling and Workforce Assignment, a mixed-integer linear programming formulation that integrates these operational decisions within a unified optimization framework. To improve computational tractability, a preprocessing procedure is developed to reduce the size of the resulting model. The proposed approach is evaluated through computational experiments on instances of increasing size and under different objective-function weight and cost configurations. Classical optimization experiments using Gurobi show that filtering method reduces the number of decision variables and constraints, while its impact on computational time varies across instances. In addition, a constrained quadratic model representation is investigated using D-Wave's hybrid quantum-classical optimization solver.

[14] arXiv:2611.18124 [pdf, html, other]
Title: Muon Attains $\mathcal{O}(\varepsilon^{-3})$ Iteration Complexity under Shuffling
Kun Huang, Shi Pu, Karl Henrik Johansson
Comments: 25 pages, 14 figures
Subjects: Optimization and Control (math.OC); Machine Learning (cs.LG)

Muon is a stochastic first-order optimizer for matrix-valued parameters. It maintains an exponential moving average of stochastic gradients in a momentum buffer with coefficient $\beta$, and orthogonalizes the buffer to obtain the update direction. Muon has been reported to accelerate and stabilize the training of large language models. In practice, it is often used with shuffling (sampling without replacement), where each epoch visits the components of a finite-sum objective in a permuted order. However, its iteration complexity in this setting remains open. We analyze Muon under shuffling, with the momentum buffer orthogonalized by the polar map. For smooth nonconvex finite-sum problems, we establish an $\mathcal{O}(\varepsilon^{-3})$ iteration complexity for finding an $\varepsilon$-stationary point. This guarantee holds for every sequence of permutations, covering random reshuffling, shuffle once, and the incremental gradient method. This complexity matches that of stochastic gradient descent under shuffling in its dependence on $\varepsilon$, and improves on Muon's $\mathcal{O}(\varepsilon^{-4})$ complexity under sampling with replacement. Our guarantee requires a problem-dependent $\beta$ that approaches one as the number of components grows. In contrast, no fixed $\beta$ guarantees convergence to stationarity for every number of components. We show this by constructing a family of smooth convex problems on which, for any fixed $\beta\in[0,1)$, Muon remains bounded away from stationarity once the number of components is sufficiently large. This failure occurs for every sequence of permutations and persists even with small stepsizes. These results suggest two practical tuning rules: if the training loss stops decreasing even after reducing the stepsize, increase $\beta$; when increasing $\beta$, reduce the stepsize accordingly.

[15] arXiv:2611.18173 [pdf, html, other]
Title: Safety Certificates for Systems under Attack via Stackelberg Set Games
Nikolaos Athanasopoulos, David Laverty, Gregory Provan, Quanyan Zhu
Subjects: Optimization and Control (math.OC); Systems and Control (eess.SY)

We study a Stackelberg set selection game for linear systems subject to bounded disturbances and additive attacks. The defender selects the vector parameter of a polytopic safe set template; the attacker, after observing this choice, selects the vector parameter of an admissible attack set. The payoff is a signed, facet-normalised invariance margin that tests invariance exactly and quantifies either the available contraction or the facet inflation needed when invariance fails. Both controlled and robust controlled invariance are treated, corresponding to different information patterns for the control input. Exploiting the structure of parameterised polyhedra, we show that within each validity region the margin is concave in the attack parameter, which reduces the follower minimisation to finitely many extreme parameters. The complete max--min game is then solved by a finite family of LPs, one per defender validity region, each enforcing all attacker regions under a common defender decision and guaranteed margin. The framework handles hard state constraints and additional costs related to the safe set size, and recovers optimal set parameters together with state and attack realisations that activate the certificate. A load frequency control example under false data injection illustrates the construction, the validity decomposition, and the two information patterns.

[16] arXiv:2611.18174 [pdf, html, other]
Title: Sample complexity and scalability of adaptive-sampling trust-region methods for zeroth-order stochastic optimization --- Part~II: a new pathwise analysis with randomized subspaces
Sara Shashaani, Coralia Cartis, Kwassi Joseph Dzahini
Subjects: Optimization and Control (math.OC)

Stochastic trust-region methods are usually analyzed in expectation or with high probability: the guarantee is about the law of the algorithm, not about the run in hand. This paper analyzes the adaptive sampling stochastic trust-region method ASTRO-DF (which is the reuse variant ASTRA-R in Part I) in random subspaces along almost every sample path. The route has one probabilistic ingredient, the accuracy of the common-random-number estimates at the adaptive sample size, uniformly over the trust region; Part~I established it by chaining, here it is obtained at the stopping time by optional stopping and converted, by the first Borel--Cantelli lemma, into an accuracy that holds from a random index on, after which the deterministic trust-region argument applies. We prove almost sure convergence and almost sure iteration and sample complexity of orders $\tilde{\mathcal{O}}(d\epsilon^{-2})$ and $\tilde{\mathcal{O}}(d^2\epsilon^{-4})$, with a subspace dimension growing logarithmically; we show that at a fixed subspace dimension the eventual-accuracy argument yields $\tilde{\mathcal{O}}(\epsilon^{-2-\mathcal{O}(1/p)})$ iterations and $\tilde{\mathcal{O}}(\epsilon^{-4-\mathcal{O}(1/p)})$ samples; we show that the random index after which the rates hold cannot be too large and has an explicit, rapidly decaying tail. Compared with the supermartingale route, the pathwise route needs no lower bound on the per-iteration probability of a good iteration, no coupling between the trust-region expansion factor and the sketch, and no moment condition on the summed squared radii.

[17] arXiv:2611.18208 [pdf, html, other]
Title: Consensus-Based Optimization of Path-Integral Free Energy for Model Predictive Control
Daisuke Inoue, Yuyang Huang, Dante Kalise
Subjects: Optimization and Control (math.OC)

We propose CBO-MPPI, a sampling-based predictive controller that applies consensus-based optimization (CBO) to the free energy underlying model predictive path integral control (MPPI). Under hard constraints, this objective rewards sampling neighborhoods containing feasible, low-cost trajectories under perturbation. Standard MPPI performs a local stochastic search by updating one nominal control sequence from Gaussian perturbations around it. In CBO-MPPI, we estimate the free energy at multiple nominal centers from local rollouts. These estimates define a shared consensus that guides drift and diffusion, and the final selected center supplies an MPPI weighted-mean output. For the associated mean-field model, we establish well-posedness and, under additional assumptions, convergence toward a free-energy minimizer. In simulations of obstacle navigation, pendulum swing-up, and planar-arm manipulation under equal rollout budgets, at high temperatures, CBO-MPPI attains lower free energy than MPPI and multi-start MPPI and maintains greater obstacle clearance than CBO-MPC.

[18] arXiv:2611.18266 [pdf, html, other]
Title: Atlas-DeePC: Learning Behavioral Atlases for Predictive Control
Dian Jin, Zhiyuan Jin, Jeremy Coulson
Comments: 25 pages, 8 figures, 4 tables. Submitted to 9th Annual Learning for Dynamics & Control Conference
Subjects: Optimization and Control (math.OC)

Data-enabled Predictive Control (DeePC) uses linear combinations of recorded input-output trajectories to compute control actions without a parametric model. For nonlinear systems, these linear combinations may not satisfy the system dynamics, and the computational cost grows with dataset size. We propose Atlas-DeePC, which approximates the behavior of a dynamical system using a collection of low-dimensional subspaces, termed a behavioral atlas, to enable computationally efficient prediction and control. The atlas is constructed offline by clustering the data and fitting a subspace to each cluster. Online, the controller selects the subspace that best matches recent measurements and solves a predictive control problem parameterized by its orthonormal basis. A graph-based reference planner complements the controller by providing intermediate references across operating regions. We illustrate Atlas-DeePC on a planar robotic manipulator, where it achieves higher success rates and substantially lower online computation times than the compared controllers based on online data selection. An additional benchmark of an autonomous racing task is also included. The project website provides code and demonstrations this https URL

[19] arXiv:2611.18302 [pdf, html, other]
Title: Integrated Topology and Redispatch Optimization of Real-World Power Transmission Systems using Branch-and-Benders-Cut
Tim Donkiewicz, Oliver Gaul, Marco Lübbecke
Subjects: Optimization and Control (math.OC)

Transmission system operators can reduce congestion management costs by jointly optimizing network topology, generator redispatch, and transformer settings. The large number of switching decisions, detailed operational constraints, and security requirements under outages make this problem computationally demanding. We present a full mixed-integer linear programming formulation that combines line switching, substation reconfiguration, generator commitment and redispatch, and transformer settings with operational requirements. We solve the problem using an exact branch-and-Benders-cut algorithm. We integrate specialized subproblem solution methods, subproblem selection, cut filtering, partial decomposition, and techniques for enforcing network connectivity. A sequential switching heuristic provides initial solutions for the exact solve. An extensive computational study on 130 synthetic instances demonstrates substantial reductions in solution time compared with standard Benders decomposition and shows how heuristic warmstarting improves solution quality, especially when more switching actions are allowed. Results on real transmission networks illustrate how joint topology and redispatch decisions identify effective switching actions that sequential approaches can miss. Selected solutions generated by our approach were presented to system operators and subsequently checked using alternating current power flow calculations. All seven discussed solutions passed a first operator assessment, indicating the practical potential of the integrated approach.

[20] arXiv:2611.18304 [pdf, html, other]
Title: Composition of Near-Planar Infrastructure Network Instances with Certified Skewness Bounds
Tim Donkiewicz
Subjects: Optimization and Control (math.OC)

Computational studies of optimization problems on infrastructure networks may require more suitable instances than are available. We propose a method to obtain additional instances by extracting and joining connected parts of existing networks, reusing their topology and parameters. Many infrastructure networks are near-planar, so added connections should preserve this property. We measure near-planarity by skewness, the minimum number of edges whose removal makes a graph planar. We derive bounds on the additional skewness caused by joining and give sufficient conditions under which none is added, and propose the corresponding strategy. We evaluate our approach on 100 sets of network parts extracted from each of PGLib-OPF and SNDlib. Within time limits, we show that no skewness is added for about half of the resulting power transmission networks and all considered telecommunication networks. Our method gives smaller mean upper bounds on added skewness than baselines that select vertices randomly, in order of decreasing degree, or using a Kamada--Kawai layout. Extraction and composition may also make the resulting optimization instances infeasible. To reduce this risk, we solve a simpler, tractable variant of the intended optimization problem during generation and restart if it is infeasible. On the generated instances, we solve security-constrained optimal transmission switching and survivable network design problems in full, demonstrating that the instances vary in computational difficulty with runtimes ranging from seconds to hours. The method thus extends the available benchmark sets with instances that inherit source data, have known skewness bounds, and cover a range of computational difficulty.

[21] arXiv:2611.18351 [pdf, other]
Title: A Parametric Active-Set Algorithm for Solving Monotone Generalized Nash Equilibrium Problems
A. Bemporad, T. Tatarenko
Comments: 29 pages, 1 figure
Subjects: Optimization and Control (math.OC)

We present a fast active-set algorithm for the solution of monotone linear-quadratic generalized Nash equilibrium problems (GNEPs) with shared affine inequality and equality constraints. The algorithm, which we call gPAQ, is an adaptation of the parametric active-set method proposed by Best (1996) and extended in qpOASES by Ferreau et al. (2014) to solve convex quadratic programs. Under strong monotonicity, we show that the same active-set strategy applies. We handle the case of merely monotone problems by using gPAQ as the inner solver within proximal-point iterations. To solve strongly monotone convex nonlinear GNEPs, we propose a Newton-type iterative scheme, which we call gPAN, which iteratively linearizes the nonlinear problem and solves the resulting LQ-GNEP subproblems using gPAQ. We show in numerical examples that the proposed algorithms are significantly faster than existing methods.

[22] arXiv:2611.18352 [pdf, html, other]
Title: Online Switching Control as Learning with Experts
Masataka Nakamura, Kaito Ito, Hideaki Ishii
Subjects: Optimization and Control (math.OC); Machine Learning (cs.LG); Systems and Control (eess.SY)

This paper addresses the problem of online switching control for discrete-time nonlinear systems where a controller is sequentially selected from a finite set of candidates. The goal is to generate an adaptive sequence of controllers to minimize the cumulative cost under unknown disturbances and cost functions. Specifically, we consider a full-information feedback setting where realized disturbance effects and cost functions are available after each selection. Using counterfactual costs computed from historical data, we reduce the online switching control problem to the experts problem with switching costs. Then, we propose an adaptive switching algorithm based on the Hedge algorithm with a switching suppression mechanism. Notably, the proposed algorithm achieves an expected regret of $O(\sqrt{T\log N})$, where $T$ is the time horizon and $N$ is the number of candidate controllers, while ensuring uniform boundedness of the state. This matches the optimal regret order for the standard experts problem.

[23] arXiv:2611.18401 [pdf, html, other]
Title: Stability Certification for Approximate Nonlinear Model Predictive Control
An Aloysius Wang, Kostas Margellos, Mark Cannon
Subjects: Optimization and Control (math.OC)

Neural-network approximations of nonlinear model predictive control can reduce online computational costs, but ensuring constraint satisfaction and closed-loop stability remains challenging. We propose a framework in which convex optimisation is used to restore feasibility under suitable local assumptions, while bounds on the corrected controller's cost suboptimality provide sufficient conditions for practical closed-loop stability. To assess these conditions over a compact set of initial states, we investigate a branch-and-bound approach based on convex relaxations of the nonlinear dynamics, with probabilistic guarantees established through scenario optimisation. The proposed approach represents a step towards combining the computational efficiency of approximate nonlinear model predictive control with systematic methods for restoring feasibility and assessing stability.

[24] arXiv:2611.18403 [pdf, html, other]
Title: Strongly Monotone Central Path Equations for Quadratic Cone Programs
Michael Cummins, Matteo Santamaria, Brendan O'Donoghue
Subjects: Optimization and Control (math.OC)

We introduce a canonical central path for the homogeneous embedding of quadratic cone programs, defined by strongly monotone equations whose modulus is the barrier parameter. We prove that the path exists uniquely, is smooth, and lies on a sphere of fixed radius. Under mild regularity assumptions, its limit points recover optimal solutions or certificates of infeasibility. Exploiting the fact that the embedding operator has a closed-form resolvent, we derive Josephy-Newton steps that retain the nonlinear embedding equations and linearize only the complementarity conditions. Each step reduces to quasi-definite linear system solves and a scalar quadratic. Using this, we derive a new predictor-corrector interior-point method with a surrogate formulation that allows the predictor and corrector to share a single sparse factorization. We provide an open-source implementation, cpo3, for linear and quadratic programs. On 675 instances from eight benchmark collections, with presolve and crossover disabled, cpo3 achieves the highest verified solve count (631) and the lowest penalized shifted geometric mean runtime among the tested open-source and commercial solvers. It is the only tested solver to solve all 138 Maros-Meszaros instances to high accuracy. Moreover, we establish Euclidean and local-norm error bounds that motivate a warm-start heuristic for the proposed algorithm, yielding up to 2.3x median runtime speedup over cold starts on perturbed Netlib and Maros-Meszaros problems.

[25] arXiv:2611.18421 [pdf, html, other]
Title: Dynamic Assortment Optimization Under Arbitrary Demand Distributions: The Uniform Price Setting
Shuo Sun, Rajan Udwani
Subjects: Optimization and Control (math.OC)

We study the joint optimization of assortment and inventory when products are identically priced and the total demand (number of customers) is uncertain. A seller chooses initial inventories before customers arrive and every customer buys her preferred product among those still in stock according to a choice model known to the seller. The objective is to maximize the (expected) total sales, subject to a capacity constraint. We give a $(1-1/e)^2$-approximation under any ranking-based choice model and a $(1-1/e-\varepsilon)$-approximation under the Markov chain choice model, where $\varepsilon$ is a tunable parameter that trades off runtime with approximation guarantee. Our guarantees hold for arbitrary demand distributions and do not require the ubiquitous Increasing Failure Rate (IFR) assumption. In fact, our results also improve on the state-of-the-art approximations for IFR demand distributions. Key to our results is the discovery of hidden submodularity in these problems and the development of efficient procedures to evaluate (or approximate) the objective values. Specifically, we uncover a novel connection to online bipartite matching (for ranking-based models) and use stochastic coupling arguments (for the Markov chain choice model) to reduce the problems to instances of constrained submodular maximization. This opens the door to applying the vast algorithmic machinery for submodular maximization to our problem setting, leading, for example, to natural generalizations beyond cardinality constraint where no approximation results were previously known.

[26] arXiv:2611.18424 [pdf, html, other]
Title: Balancing-based model reduction for output-constrained optimal control of switched linear systems
Michael Kartmann, Mattia Manucci, Benjamin Unger, Stefan Volkwein
Comments: 24 pages
Subjects: Optimization and Control (math.OC); Numerical Analysis (math.NA)

We develop a balanced-truncation (BT) framework for model-order reduction of optimal control problems governed by switched linear systems, with quadratic tracking costs, convex input constraints, and Moreau--Yosida-regularized output constraints. For general Petrov--Galerkin reduced models, we derive an a posteriori bound for the error in the optimal value function, and combine it with BT-based estimates of the dynamical-system reduction error. We show that BT preserves the duality structure underlying the first-order optimality conditions, making first-reduce-then-optimize and first-optimize-then-reduce equivalent. Consequently, reducing the forward system also reduces the associated optimality system, leading to a certified, online-efficient, and online-adaptive approximation of the parameterized solution operator. The reduced model can therefore be reused across different regularization parameters, constraint sets, switching signals, and reference trajectories.

[27] arXiv:2611.18537 [pdf, html, other]
Title: Valid Inequalities for Bilevel 0-1 Shortest-Path Problems
Bruno M. Pacheco, Margarida Carvalho, Kim Yu
Subjects: Optimization and Control (math.OC)

We characterize valid inequalities for bilevel programs in which the leader makes binary decisions, and the follower solves a shortest-path problem. We exploit a dualize-and-combine reformulation and show that even with ideal formulations for the leader and the follower primal solution sets, fractional solutions may still appear in the continuous relaxation. From that reformulation, we identify that the interaction between the leader variables and the follower dual variables, which we capture through a leader-potential set, is the key structure to be convexified to guarantee that a continuous relaxation recovers the optimal value of the original problem, closing the objective gap. Furthermore, we characterize the valid inequalities for the leader-potential set through network flows, and identify a family of inequalities we name root cuts that suffice to recover the optimal value. Complementary to the structural results, we prove that the cut generation problem is NP-hard even with partially fixed coefficients and a network with two nodes and two parallel arcs. Motivated by the complexity result, we propose a constructive method to fix coefficients in the valid inequalities while preserving separation guarantees, by exploiting the network structure.

[28] arXiv:2611.18547 [pdf, html, other]
Title: Polynomial-Time Solvability of the Multi-Dimensional Slope Change Problem
Apoorva Narula, Santanu S. Dey
Subjects: Optimization and Control (math.OC)

Slope change detection seeks to identify changes in the trends of time-ordered signals. We study the problem of fitting a multidimensional time series by continuous piecewise-linear functions with a shared continuous breakpoint and unrestricted slopes. We consider both the dense setting, in which every dimension may change slope, and the sparse setting, in which at most $k$ dimensions may change slope. Although the resulting optimization problems are nonconvex mixed integer nonlinear programs and may have irrational optimal solutions even for rational data, we prove that all six variants, corresponding to the $\ell_1$, $\ell_2$, and $\ell_\infty$ losses in both settings, are exactly solvable in polynomial time. For least-squares loss, we obtain efficient algorithms requiring $O(JT)$ operations in the dense setting and $O(J^2T\log J)$ operations in the sparse setting, where $T$ is the number of time points and $J$ is the number of dimensions. We implement these least-squares algorithms and evaluate them on simulated data and three real-world time series, namely, surface temperatures (ERA5), hand-tracking data from a mixed-reality headset (HoloAssist), and turbofan engine sensor data (NASA C-MAPSS). The dense algorithm solves instances with up to $J=10^5$ and $T=100$, that is, $10^7$ observations, in less than $0.1$ seconds on a laptop, while the sparse algorithm solves instances with about $10^3$ dimensions and up to $T=240$ in under a second. Compared with a mixed-integer nonlinear programming (MINLP) formulation solved by Gurobi, our algorithms are several orders of magnitude faster (approximately $8.9 \times 10^3$ to $3.4 \times 10^6$ times) on instances where Gurobi terminates within its default optimality tolerance of $0.01\%$ optimality gap. On larger instances, Gurobi frequently reaches a 10,000-second time limit with substantial optimality gaps.

Cross submissions (showing 12 of 12 entries)

[29] arXiv:2611.17459 (cross-list from cs.LG) [pdf, html, other]
Title: Classification with Large-Step Gradient Descent: Unifying Self-Stabilization and Convergence Rates
Salma Tarmoun, Lachlan Ewen MacDonald, Konstantinos Emmanouilidis, Ziqing Xu, René Vidal
Subjects: Machine Learning (cs.LG); Optimization and Control (math.OC)

Neural networks are often trained with gradient descent (GD) using step sizes that exceed classical stability thresholds, a regime where standard convergence theory no longer applies. Despite exhibiting non-monotonic loss and oscillatory dynamics, this regime frequently leads to faster convergence and better generalization. Existing theoretical perspectives offer complementary but incomplete insights: one identifies a self-stabilization mechanism that drives GD toward flat regions but does not yield convergence rates; another derives rigorous rates for specific losses but does not explain the underlying stabilization mechanism; and a third unifies these views for overparameterized least squares, but relies on the existence of a minimizer manifold, which is absent in classification settings. In this work, we extend this unified geometric perspective to exponential-tail losses, including logistic and cross-entropy losses in separable linear classification, where the loss infimum is not attained at any finite parameter. We identify a curvature-defined reference subspace that replaces the role of the minimizer manifold, yielding a coupled dynamical system in which orthogonal oscillations are damped by parallel progress along directions of decreasing sharpness. Under regularity conditions, which we verify for logistic and multiclass cross-entropy losses on symmetric data models, we derive explicit bounds for the iterates, not just the loss, covering both the unstable and stable regimes, and revealing a three-phase damping structure in the latter. Our analysis further uncovers a transient implicit bias induced by large step sizes: although GD converges asymptotically to the max-margin direction, large steps can leave persistent orthogonal residuals and, in certain geometries, amplify them, affecting finite-time alignment with the max-margin direction.

[30] arXiv:2611.17484 (cross-list from cs.LG) [pdf, html, other]
Title: SURFACE: Preference Distribution Matching on Pareto Fronts via Scalarization Steering
Liuyuan Jiang, Prasanna Sattigeri, Lisha Chen
Subjects: Machine Learning (cs.LG); Optimization and Control (math.OC); Machine Learning (stat.ML)

Constructing a finite Pareto representation requires allocating a limited number of solutions across the Pareto front (PF). A desired preference distribution specifies this allocation by assigning more solutions to some PF regions and fewer to others. Scalarization generates Pareto solutions by varying objective weights and solving single-objective problems. However, a distribution of scalarization weights generally does not translate into the same distribution of Pareto solutions along the PF, because the weight-to-solution map is nonlinear and unknown in this http URL account for this distortion, we introduce SURFACE (Scalarization Updates for Redistributing Finite Approximations with Chart Estimation}), which adaptively updates scalarization weights to match a desired PF distribution. SURFACE estimates how each region of the weight grid corresponds to an area of the PF from the current scalarized solutions, then redistributes the weights accordingly. The method supports parallel, warm-started, inexact optimization with standard single-objective solvers. Theoretically, we establish that, under the stated regularity conditions, SURFACE achieves linear convergence of the preference distribution mismatch to a finite-resolution error floor of $\mathcal{O}(N^{-2/(M-1)})$, where $N$ is the number of adaptive weights and $M$ is the number of objectives. Empirically, experiments across multiple benchmarks demonstrate improved target-mass balance and PF coverage compared with established baselines.

[31] arXiv:2611.17504 (cross-list from eess.SY) [pdf, html, other]
Title: Learning monotone game-theoretic linear MPC
Alberto Bemporad, Tatiana Tatarenko
Comments: 15 pages, 1 figure
Subjects: Systems and Control (eess.SY); Optimization and Control (math.OC)

We propose a method to learn a game-theoretic model predictive control (GT-MPC) law by imitation from input and output data of a running multi-agent system. By interpreting the recorded inputs as the first moves of the equilibria of a game played by the agents, we learn the GT-MPC weights under hard input constraints and soft output constraints, without requiring the agents to reveal their objectives. We also impose that the resulting generalized Nash equilibrium (GNE) problem is strongly monotone, which ensures the existence and uniqueness of the equilibrium and, hence, that the learned GT-MPC law is well defined and can be evaluated efficiently online by variational GNE-seeking algorithms for linear-quadratic games. Moreover, strong monotonicity implies that the equilibrium is almost everywhere differentiable with respect to the references and the weights, which enables automatic differentiation for gradient-based optimization. In addition, we promote closed-loop stability of the unconstrained equilibrium feedback by incorporating a penalty that involves a jointly learned quadratic Lyapunov function. Learning a GT-MPC formulation enables us to explain the agents' objectives, reproduce their behavior in simulation in new scenarios, optimize their set-points, design mechanisms, and enforce additional constraints. A numerical example illustrates the approach.

[32] arXiv:2611.17902 (cross-list from q-fin.MF) [pdf, html, other]
Title: Well-Posedness of Semimartingale Optimal Transport with Partial Marginal Constraints for Model Calibration
Jean-François Chassagneux, Stéphane Crépey, Grégoire Loeper, Justin Ruelland
Subjects: Mathematical Finance (q-fin.MF); Optimization and Control (math.OC); Probability (math.PR)

We study the well-posedness and stability of a semimartingale optimal transport problem with partial marginal constraints at several maturities, motivated by the calibration of financial models to finitely many option prices. We connect its probabilistic and Eulerian formulations within the Benamou--Brenier framework. Under convex-duality assumptions, we establish strong duality for every price vector in the extended-value sense and existence of a primal optimizer whenever the value is finite. Under additional PDE assumptions, we establish dual attainment on the relative interior of the admissible price set, together with uniqueness of the primal optimizer. Under further non-degeneracy assumptions, we show that this price set has nonempty interior, on which the dual optimizer is unique and depends locally Lipschitz continuously on the prices. The analysis combines convex methods with regularity and local strong convexity results for Hamilton--Jacobi--Bellman solution maps established in a companion paper. For a fixed deterministic initial state, we also prove continuity of the optimal law on path space in the adapted Wasserstein distance of order two as prices vary within this interior. This stability concerns the optimizer over the full admissible class of semimartingale laws, allowing characteristics that depend on the past. The framework accommodates bounded uniformly continuous payoffs, including European puts in log-price coordinates, without payoff regularization or spatial truncation.

[33] arXiv:2611.18094 (cross-list from eess.SY) [pdf, html, other]
Title: Controllability-Informed Distributed Electric Propulsion Aircraft Design Using Probabilistic Attainable Moment Sets
Umut Demir, Torbjørn Cunis, Aamir Ahmad, Walter Fichter
Subjects: Systems and Control (eess.SY); Optimization and Control (math.OC)

Integrated aircraft design combines models of aerodynamics, propulsion, mass, and mission performance in one design loop, so that the disciplines can be traded against each other from the earliest stage. For configurations with distributed electric propulsion (DEP), part of the control authority comes from the propulsion system itself, which directly affects the required size of the tail and should therefore be known while the aircraft is being sized. This paper presents a controllability measure that is evaluated for every design in the loop and returned to it as a design signal. The attainable moment set of each candidate design is computed from the surrugate models of the loop and it is compared with a required moment set. Since the low-fidelity surrugate models are uncertain, the comparison is formulated with probabilistic attainable moment sets: the uncertainty is propagated to the boundary of the set, which yields a margin for every direction in the moment space and the probability of violating the requirement. The design loop thereby informed whether a configuration provides sufficient control authority, on which axes it falls short, and how this changes with the design variables. The approach is demonstrated on a DEP regional aircraft, where the signal is used to assess the size of the vertical tail.

[34] arXiv:2611.18252 (cross-list from cs.LG) [pdf, html, other]
Title: ParetoM$^3$: Learning on the Pareto Set under Preference Guidance via Min-Max-Min Optimization
Pei Tang, Songtao Lu
Comments: Accepted at the 40th Annual Conference on Neural Information Processing Systems (NeurIPS 2026)
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Optimization and Control (math.OC)

Learning across multiple inherently conflicting objectives requires selecting desirable trade-offs from the Pareto set. In this work, we study preference-guided multi-objective optimization, where a predefined preference function is minimized over the weak Pareto set of multiple nonconvex objectives. Leveraging a merit function that vanishes on the weak Pareto set, we propose a penalty-based min-max-min formulation, termed the M$^3$ problem, which unifies preference optimization and the enforcement of weak Pareto optimality in a single objective. We further develop ParetoM$^3$, a single-loop first-order algorithm for computing stationary solutions of the resulting M$^3$ problem. Under a local Kurdyka-Lojasiewicz assumption, we prove that ParetoM$^3$ finds an $\epsilon$-stationary point of the M$^3$ problem that is also approximately weakly Pareto optimal for the original multi-objective problem when the penalty parameter is sufficiently large, with explicit non-asymptotic convergence rates. We also show that this stationarity notion recovers existing optimality criteria, including approximate preference stationarity and approximate Karush-Kuhn-Tucker conditions, under the same assumptions used in prior works. Experiments on a synthetic benchmark, image classification, regression, and math reasoning demonstrate that ParetoM$^3$ reliably attains preferred weak Pareto optimal solutions while achieving strong practical performance. Code is available at this https URL.

[35] arXiv:2611.18313 (cross-list from eess.SY) [pdf, html, other]
Title: Hierarchical Interactions in Mobility Networks: A Model-Free Feedback Approach
Mingjia He, Zhiyu He, Jan Ghadamian, Florian Dörfler, Emilio Frazzoli, Gioele Zardini
Comments: 12 pages
Subjects: Systems and Control (eess.SY); Optimization and Control (math.OC)

Urban mobility systems are large-scale networked systems whose performance emerges from hierarchically coupled decisions of public authorities, competing mobility operators, and travelers. Existing approaches to policy design either collapse one of these levels or assume that the authority can model and differentiate through the equilibrium responses of operators and travelers. In this paper, we model policy design as a tri-level game on a multimodal network, in which a municipality sets taxes, licenses, and integration subsidies, a public transport and a taxi operator compete on prices and service supply, and heterogeneous travelers reach a stochastic user equilibrium. We prove that the traveler equilibrium is unique and continuously differentiable in the operator strategies, that the variational inequality characterizing operator equilibria admits a solution for every policy, and that the resulting tri-level game admits a Stackelberg equilibrium. Building on this structure, we design a model-free hierarchical controller in which the municipality and the operators update their decisions from zeroth-order evaluations of observed equilibrium outcomes. On the Zurich multimodal network, operator equilibria computed from different initializations differ by a median of 0.3% in the municipal objective. On a grid of 3,125 policies, only those that subsidize transfers between public transport and taxis benefit both operators, and 24 of them also lower travel costs without degrading the municipal objective. Finally, the controller matches the best policies of three black-box optimizers and terminates at the most consistent policies across runs.

[36] arXiv:2611.18368 (cross-list from math.AP) [pdf, html, other]
Title: The BCC Lattice Maximizes the First Positive Neumann Eigenvalue among 3D Lattice Voronoi Cells
Kaixin Deng, Senping Luo
Subjects: Analysis of PDEs (math.AP); Mathematical Physics (math-ph); Optimization and Control (math.OC); Spectral Theory (math.SP)

It is shown that, among three-dimensional lattices of fixed covolume, the body-centered cubic lattice maximizes the first positive Neumann eigenvalue of its Voronoi cell uniquely, up to orthogonal transformations.

[37] arXiv:2611.18447 (cross-list from quant-ph) [pdf, html, other]
Title: Nesting in additive manufacturing using quantum annealing
Xiangqian Wu, Jannik Gayer, Marcel Wagner, Jan C. Aurich
Subjects: Quantum Physics (quant-ph); Optimization and Control (math.OC)

Nesting in powder bed fusion (PBF), an additive manufacturing (AM) process, determines the positions and orientations of parts within the build space and influences build height and build time. As the number of parts and feasible placements increases, the nesting problem becomes computationally demanding. Exact approaches are limited to small problem sizes as computational effort increases rapidly with the problem size, while heuristic and metaheuristic approaches involve a trade-off between solution quality and computation time. Quantum annealing (QA) offers an alternative for solving combinatorial optimization problems, but its application to nesting in AM remains largely unexplored. This paper presents a QA-based approach for nesting in AM to investigate its feasibility and its scaling behavior across three problem sizes. An iterative binary quadratic model based approach (IBQM) is developed to minimize build height while considering the geometric characteristics of AM parts. Geometric abstraction, iterative decomposition, and variable pruning are applied to address the computational complexity of the nesting problem. The approach is evaluated using three cases with increasing problem sizes and compared to a classical approach combining tabu search and simulated annealing (TBSA). Compared with TBSA, IBQM reduces the build height of the original parts by up to 13.32% and the computation time by up to 74.27%.

[38] arXiv:2611.18455 (cross-list from stat.ML) [pdf, html, other]
Title: Mesh-Free Weak-Form Regression for Learning Interaction Kernels from Particle Trajectories
Daisuke Inoue, Dante Kalise
Comments: 47 pages, 8 figures
Subjects: Machine Learning (stat.ML); Optimization and Control (math.OC)

We propose mesh-free weak-form regression for learning interaction kernels from noisy particle trajectories. Our method uses smooth test functions to avoid differentiating noisy data and empirical particle averages to avoid density reconstruction and spatial grids. We then fit sparse combinations of candidate functions, evaluating interaction terms on sampled particle pairs to avoid all-pair quadratic cost. By transferring identifiability from the mean-field limit, we prove high-probability coefficient error bounds that separate finite-particle, time-discretization, observation-noise, and pair-sampling errors, without grid-resolution or density-estimation terms. We evaluate one-dimensional (1D) Newtonian-repulsion, 2D predator-prey, and 10D distance-localized experiments. The numerical results show trends consistent with the theoretical coefficient-error bounds. Comparisons of weak and forward-difference formulations with LASSO, Ridge, and OLS show that weak LASSO gives the smallest coefficient and drift errors in the three nominal experiments.

[39] arXiv:2611.18480 (cross-list from stat.ML) [pdf, html, other]
Title: Bregman geometry in regular exponential families: dual representations and relative smoothness
Mame Diarra Fall, Éric Barat, Nicolas Dobigeon
Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG); Optimization and Control (math.OC)

Regular exponential-family distributions are ubiquitous in statistics due to their appealing and convenient properties. Notably, they can be expressed in terms of Bregman divergences, revealing fundamental connections between statistics, convex analysis, and information geometry. However, this geometric structure has not been systematically exploited to characterize these distributions with respect to their expectation parametrization. Building on a projective characterization, we establish a general construction of equivalent dual Bregman representations, together with normalization and uniqueness results. These representations express log-densities as ordinary Bregman divergences in dual coordinate systems, providing geometric characterizations of the parameter space. So-called $\beta$- and $\alpha$-scaled dual representations emerge as particular instances of this framework. We further derive dual Bregman representations of conjugate priors and establish that they preserve conjugacy. Applications to classical distributions illustrate the geometric structures underlying seemingly different statistical models. Moreover, we show that these representations provide natural reference geometries for establishing relative smoothness of the primal Bregman representation of the distribution, with explicit bounds that are sharp in the scalar and separable settings. An illustrative application to the Poisson distribution sheds new light on several works in the literature, providing complementary geometric justifications for reference geometries while deriving existing algorithms.

[40] arXiv:2611.18502 (cross-list from eess.SY) [pdf, html, other]
Title: EPIGRIDS: Documentation and Validation of Seven Open Synthetic Transmission Grid Cases
Jonathan M. Snodgrass, Scott Greene, Thomas J. Overbye, Christopher L. DeMarco, Bernard C. Lesieutre
Comments: 10 pages, 2 figures, 6 tables. Submitted to the IEEE Open Access Journal of Power and Energy. Cases: this https URL load series: this https URL construction method: arXiv:2611.16222
Subjects: Systems and Control (eess.SY); Optimization and Control (math.OC)

This paper documents and checks EPIGRIDS, seven open synthetic transmission grid cases from four independently built networks, made with an optimization-based construction method instead of the graph-statistics method behind most large public synthetic grids. The cases range from 250 to 78,484 buses, with geolocated substations, voltages up to 765 kV, and line parameters taken from conductor data. Each ships under an open license in three common power flow formats that agree at the solved operating point, and the six larger cases have a companion hourly, bus-level load series for one year. The Texas, Florida, and Michigan, Ohio, and Indiana cases are sub-networks of the Eastern case. Every case solves an alternating-current power flow from its stored operating point. With parallel circuits merged, the Eastern and Western cases are 9 and 16% denser than the real interconnections, while graph-statistics cases on the same footprints lie within 6% and meet more published realism criteria. A full single-outage contingency analysis completes on every case, and in an optimal power flow benchmark every case except the New England base case and one of its 12 operating points solves in polar form.

Replacement submissions (showing 27 of 27 entries)

[41] arXiv:2210.13072 (replaced) [pdf, html, other]
Title: Demystifying the characterization of SDP matrices in mathematical programming
Daniel Porumbel
Comments: 100 pages
Subjects: Optimization and Control (math.OC)

This manuscript was written because I found no other introduction to SDP programming that targets the same audience. A first difference compared to other existing introductions to SDP is that this work comes out of a mind that was itself struggling to understand. This may seem to be only a weakness, but, paradoxically, it is both a weakness and a strength. First, I did not try to overpower the reader, but I tried to minimize the distance between the author and the reader as much as possible, even hoping to achieve a small level of mutual empathy. This enabled me avoid a quite common pitfall: many long-acknowledged experts tend to forget the difficulties of beginners. Other experts try to make all proofs as short as possible and to dismiss as unimportant certain key results they have seen thousands of time in their career. I also avoided this, even if I did shorten a few proofs when I revised this manuscript two years after it was first written. However, I also kept certain proofs that seem longer than necessary because I feel they offer more insight; an important goal is to capture the "spirit" of each proven result instead of reducing it to a flow of formulae.
The very first key step towards mastering SDP programming is to get full insight into the eigen-decomposition of real symmetric matrices. It is enough to see the way many other introductions to SDP programming address this eigen-decomposition to understand that their target audience is different from mine. They often list the eigen-decomposition without proof, while I give two proofs to really familiarize the reader with it.

[42] arXiv:2211.03997 (replaced) [pdf, html, other]
Title: Online Nonconvex Decision Making with Approximate Oracles
Rui Chen, Oktay Gunluk, Andrea Lodi, Guanyi Wang
Subjects: Optimization and Control (math.OC); Data Structures and Algorithms (cs.DS)

We study online decision making with possibly nonconvex (for example, combinatorial or integer) local feasible sets and convex long-term constraints. At each period, a decision maker selects a local action to maximize cumulative reward while ensuring that the sequence of actions satisfies the global constraints in the long term. It is known that $O(\sqrt{T})$ long-term constraint violation and $O(\sqrt{T})$ reward regret are achievable if given access to a local optimization oracle that exactly solves optimization problems over each local feasible set. Instead of requiring exact optimization over each local feasible set, we assume access to an approximate local optimization oracle that permits optimization errors quantified by two parameters $(\delta_+,\delta_\times)$. We develop an approximate online dual mirror descent algorithm and show that its guarantees depend on a feasibility slack parameter $\underline d$, which measures the compatibility between local feasible sets and the long-term constraint. Under the standard random permutation model, when $\delta_+=O(1)$, $\delta_\times<\underline d$ and $\underline d-\delta_\times=\Omega(1)$, the algorithm achieves $O(\sqrt T)$ long-term constraint violation and reward regret up to $O\bigl(\sqrt T+\delta_\times T\bigr)+\delta_+ T$. We also show a matching limitation: when $\delta_+=0$ and $\delta_\times>\underline d$, no online algorithm relying on such approximate oracle outputs can guarantee sublinear long-term constraint violation in general.

[43] arXiv:2309.10242 (replaced) [pdf, html, other]
Title: Reinforcement Learning for optimal dividend problem under diffusion model
Lihua Bai, Thejani Gamage, Jin Ma, Gaozhan Wang
Comments: In this version the numerical scheme and the experiments are reconstructed
Subjects: Optimization and Control (math.OC)

In this paper, we study the optimal dividend problem under the continuous time diffusion model with the bounded dividend rate from the Reinforcement Learning (RL) perspective. Unlike the standard literature, our main focus will be on numerical algorithms that allow part or all of the system parameters to be unspecified so that the optimal control cannot be explicitly determined. Following the RL literature we introduce the entropy-regularized exploratory control problem, which randomizes the control actions and balances the levels of exploitation and exploration, and carry out a theoretical analysis of the associated Policy Improvement (PI) and Policy Evaluation (PE) devices and the corresponding sequence of the approximating optimal strategies. Specifically, our algorithm will be based on two independent neural networks that approximate the value function and its derivative simultaneously. Such an algorithm, to the best of our knowledge, is new in the context of the optimal dividend problems, and can be effective even for the situation when the premium and/or interest rate is state dependent, hence beyond reach of the standard statistical methods. Some numerical experiments are presented to empirically demonstrate the effectiveness of our RL algorithm.

[44] arXiv:2501.04151 (replaced) [pdf, html, other]
Title: Efficient LP warmstarting for linear modifications of the constraint matrix
Guillaume Derval, Bardhyl Miftari, Damien Ernst, Quentin Louveaux
Subjects: Optimization and Control (math.OC); Computational Complexity (cs.CC)

We consider the problem of computing the optimal solution and objective value of a linear program whose constraint matrix changes linearly with a parameter, for a predefined set of values of this parameter. Based on the information given by a precomputed basis, we present three efficient LP warm-starting algorithms, based on the eigenvalue decomposition, the Schur decomposition and the Hessenberg decomposition, to evaluate the optimal solution and optimal objective value of these problems. The three algorithms have an overall complexity of $O(pm^2+pmn)$ where $m$ (resp. $n$) is the number of constraints (resp. variables) of the original problem and $p$ the number of parameter values after an initial preprocessing step. We also provide theorems related to the optimality conditions to verify when a basis is still optimal. When the perturbation of the basis matrix has a low rank $\rho$, as is often the case in practice, we then reduce all the computations to a $\rho\times\rho$ matrix built with $\rho$ solves with the factorization of the basis maintained by the simplex solver; with it, we compute, in each direction, an interval on which the basis is guaranteed to remain optimal, together with a bound on the objective. In practice, these intervals are close to the exact ones, and they remain useful for degenerate bases. We also propose a matrix-free method based on shifted Krylov subspaces. Numerical experiments on 398 instances built from real models, with up to 349,120 constraints, assess the accuracy, the cost and the range of applicability of all the methods. On the interval on which a basis remains optimal, once this interval is known, the low-rank and Krylov methods compute the optimal objective value 117 to 304 times faster than a warm start of the solver, in median.

[45] arXiv:2504.11141 (replaced) [pdf, html, other]
Title: A definition of the background state of the atmosphere using optimal transport
Charlie Egan, John Methven, David P. Bourne, Mike J. P. Cullen
Subjects: Optimization and Control (math.OC)

The dynamics of atmospheric disturbances are often described in terms of displacements of air parcels relative to their locations in a notional background state. Modified Lagrangian Mean (MLM) states have been proposed by M. E. McIntyre using the Lagrangian conserved variables potential vorticity and potential temperature to label air masses in the instantaneous 3-D atmospheric state, and then rearranging air masses sandwiched between potential vorticity contours in potential temperature layers so that they are zonally symmetric. Methven and Berrisford used conservation of Kelvin's circulation to define the distribution of MLM state mass in zonal angular momentum ($z$) and potential temperature ($\theta$) coordinates given 3-D global data. We prove that for any such mass distribution in $(z, \theta)$ there exists a mapping to a unique mass distribution in geophysical coordinates (latitude and pressure) which minimises energy and completely defines the MLM state in a single hemisphere. Since the state is an energy minimum, it is suitable for quantification of finite amplitude disturbances and examining atmospheric instability. This state is obtained by solving a free surface problem, which we frame as the minimisation of an optimal transport cost over a class of source measures. The solution consists of a source measure, encoding surface pressure, and an optimal transport map, connecting the distribution of mass in geophysical coordinates to the known distribution of mass in $(z, \theta)$. We show that this problem reduces to an optimal transport problem with a known source measure, which has a numerically feasible discretisation, and we present a numerical solution using atmospheric reanalysis data. Additionally, our results hold for a large class of cost functions, and generalise analogous results on free surface variants of the semi-geostrophic equations.

[46] arXiv:2505.17965 (replaced) [pdf, html, other]
Title: Bias-Optimal Bounds for Stochastic Gradient Descent
Daniel Cortild, Lucas Ketels, Juan Peypouquet, Guillaume Garrigos
Comments: 30 pages. Under review
Subjects: Optimization and Control (math.OC); Machine Learning (cs.LG); Machine Learning (stat.ML)

The non-asymptotic analysis of Stochastic Gradient Descent (SGD) typically yields bounds that decompose into a bias term and a variance term. In this work, we focus on the bias component and study the extent to which SGD can match the optimal convergence behavior of deterministic gradient descent. Assuming only (strong) convexity and smoothness of the objective, we derive new bounds for the full range of constant step-sizes $\gamma L \in (0,2)$. This includes critical and large step-size regimes that were previously unexplored without additional variance assumptions, in both the convex and strongly convex cases. Those bounds are bias-optimal, in the sense that the bias term coincides with the worst-case rate of gradient descent, for all non-critical step-sizes. For critical step-sizes, we show that no bounds can be obtained with an optimal bias term, but that this optimal bias can be approximated arbitrarily closely. As an immediate application, we provide new complexity bounds for the Stochastic Proximal algorithm, without the need of regularity assumptions.

[47] arXiv:2506.04948 (replaced) [pdf, other]
Title: Entropic smoothing and gradient approximation for Wasserstein distributionally robust optimization
Tam Le (LPSM (UMR\_8001))
Subjects: Optimization and Control (math.OC); Machine Learning (stat.ML)

Wasserstein distributionally robust optimization offers a framework for model fitting in machine learning under potential shifts in the data distribution. We study a regularized variant of this problem in which entropic smoothing produces a sampled approximation of the original objective. We establish convergence of the approximate gradients to subgradients of the unregularized objective as the regularization parameter vanishes, enabling convergence guarantees for stochastic gradient methods. We obtain qualitative convergence results under general assumptions and provide convergence rates under additional regularity. In particular, we prove rates for the convergence of the unregularized objective values, up to sampling errors, when the regularization level is decreased across iterations. Our analysis also yields results of independent interest on approximating the subdifferentials of maximum functions by gradients of their smooth approximations, together with growth-based lower bounds on optimal dual variables in Wasserstein distributionally robust optimization.

[48] arXiv:2509.03251 (replaced) [pdf, html, other]
Title: Multi-period Asset-liability Management with Reinforcement Learning in a Regime-Switching Market
Zhongqin Gao, Ping Chen, Xun Li, Yan Lv, Wenhao Zhang
Comments: 25 pages, 3 figures
Subjects: Optimization and Control (math.OC); Probability (math.PR)

This paper explores the mean-variance portfolio selection problem in a multi-period financial market characterized by regime-switching dynamics and uncontrollable liabilities. To address the uncertainty in the decision-making process within the financial market, we incorporate reinforcement learning (RL) techniques. Specifically, the study examines an exploratory mean-variance (EMV) framework where investors aim to minimize risk while maximizing returns under incomplete market information, influenced by shifting economic regimes. The market model includes risk-free and risky assets, with liability dynamics driven by a Markov regime-switching process. To align with real-world scenarios where financial decisions are made over discrete time periods, we adopt a multi-period dynamic model. We present an optimal portfolio strategy derived using RL techniques that adapt to these market conditions. The proposed solution addresses the inherent time inconsistency in classical mean-variance models by integrating a pre-committed strategy formulation. Furthermore, we incorporate partial market observability, employing stochastic filtering techniques to estimate unobservable market states. Numerical simulations and empirical tests on real financial data demonstrate that our method achieves superior returns, lower risk, and faster convergence compared to traditional models. These findings highlight the robustness and adaptability of our RL-based solution in dynamic and complex financial environments.

[49] arXiv:2510.08782 (replaced) [pdf, html, other]
Title: A generalized alternating NGMRES method for PDE-constrained optimization problems governed by transport equations
Yunhui He, Andreas Mang
Comments: 34 pages, 3 figures, 14 tables
Subjects: Optimization and Control (math.OC)

We propose a generalized alternating nonlinear generalized minimal residual method (GA-NGMRES) to accelerate first-order optimization schemes for PDE-constrained optimization problems governed by transport equations. Interpreting
the update rule of a gradient descent scheme in the Hilbert space induced by the regularization model, i.e., a Sobolev gradient descent, as a fixed-point (FP) iteration, GA-NGMRES periodically alternates NGMRES updates with FP
steps. It requires only first-order derivatives. We consider advection, mass-preserving transport, and incompressible flows as PDE models; variational regularization of the control variable promotes diffeomorphic transport maps,
which we verify numerically. With a line search safeguard, GA-NGMRES converges globally. Since its FP map need not be contractive, we extend a recent analysis of NGMRES for noncontractive iterations to the alternating scheme. We
compare GA-NGMRES to this Sobolev gradient descent, Anderson acceleration, a limited-memory BFGS method in the same Hilbert space, and Gauss-Newton-Krylov methods on real-world and synthetic data. GA-NGMRES converges where the
unaccelerated scheme does not and outperforms Anderson acceleration. On our problems, it does not, in general, outperform the (quasi-)Newton methods, but it closes much of the gap to them; which method is fastest depends on the
preconditioner, the regularization parameter, the data, and the stopping tolerance. GA-NGMRES is a simple and robust option when moderately accurate solutions suffice.

[50] arXiv:2602.06567 (replaced) [pdf, html, other]
Title: Markov Decision Processes of the Third Kind: Learning Distributions by Policy Gradient Descent
Nicole Bäuerle, Athanasios Vasileiadis
Subjects: Optimization and Control (math.OC)

The goal of this paper is to analyze distributional Markov Decision Processes as a class of control problems in which the objective is to learn policies that steer the distribution of a cumulative reward toward a prescribed target law, rather than optimizing an expected value or a risk functional. To solve the resulting distributional control problem, we propose a policy-gradient algorithm based on neural-network parameterizations of randomized Markov policies, defined on an augmented state space and a sample-based evaluation of the characteristic-function loss. Under mild regularity and growth assumptions, we prove convergence of the algorithm to stationary points using stochastic approximation techniques. Several numerical experiments illustrate the ability of the method to match complex target distributions, recover classical optimal policies when they exist, and reveal intrinsic non-uniqueness phenomena specific to distributional control.

[51] arXiv:2604.02058 (replaced) [pdf, html, other]
Title: Faster Symmetric Rendezvous on Four or More Locations
Javier Cembrano, Felix Fischer, Max Klimm
Subjects: Optimization and Control (math.OC); Computer Science and Game Theory (cs.GT)

In the symmetric rendezvous problem, two players follow the same (randomized) strategy to visit one of $n$ locations in each time step $t=0,1,2,\dots$. Their goal is to minimize the expected time until they visit the same location and thus meet. A canonical strategy due to Anderson and Weber (J. Appl. Probab., 1990) is known to be optimal for $n=2$ and $n=3$ (Weber, Math. Oper. Res. 2012), but whether it remains optimal for larger values of $n$ has been an open question since 1990. We show that it does not remain optimal: for any finite $n\geq 4$, we construct an explicit symmetric strategy that achieves a strictly smaller expected meeting time than the Anderson-Weber strategy.
In the Anderson-Weber strategy players stay at a dedicated home location for $n-1$ steps with a certain probability $\theta$ and with the remaining probability tour all non-home locations in a random order. Our improving strategy introduces carefully chosen correlations between consecutive tours of the non-home locations. The construction is uniform in $n$ and is guided by a graph-theoretic view in which tours correspond to permutations and meetings to edges in the complement of the derangement graph. By exploiting the clique structure of this graph we obtain a correlated strategy that improves on the Anderson-Weber strategy. For $n=4$, we give an exact expression for the improvement; for any $n\geq 5$, we obtain a lower bound on the expected improvement of $\frac{483(1-\theta)^6}{(n-1)^8}$, where $\theta \in [0,1)$ is the probability of staying at the home location. The graph-theoretic framework we introduce may be useful more widely in the design and analysis of correlated strategies for rendezvous.

[52] arXiv:2605.02754 (replaced) [pdf, html, other]
Title: Transferability of Error Bounds and the Kurdyka-Łojasiewicz Property under $C^1$ Partial Smoothness
Hanju Wu, Ching-pei Lee, Yue Xie
Subjects: Optimization and Control (math.OC)

This work studies how a generalized error bound (GEB) property in the ambient Euclidean space transfers to the active manifold under $C^1$ partial smoothness. The GEB generalizes the standard error bound, which upper bounds the distance to a subset of critical points using first-order stationarity residuals, by allowing the distance to be composed with a gauge function. In the ambient space, these residuals are measured by the subdifferential or the slope of the original function, while on the manifold they are measured by the Riemannian gradient or the slope of the restricted function. We develop complementary geometric and metric analyses to show that the ambient and manifold GEBs are equivalent with the same gauge function. The geometric argument shows that the projection of the subdifferential onto the tangent space coincides with the Riemannian gradient, while the metric approach combines slope estimates, identifiability, and a new uniform linear-growth result for $C^1$ manifolds. We further extend our analysis to establish the analogous equivalence for the Kurdyka--Łojasiewicz (KL) property, with the same desingularizer, and separately characterize their connections with the GEB. Finally, for regularized optimization, we relate subdifferential and proximal error bounds, thereby placing the previously established equivalence between proximal and manifold error bounds for $\ell_1$-regularization within a broader framework.

[53] arXiv:2605.29371 (replaced) [pdf, html, other]
Title: Kernel-based potential mean-field games with unbiased random Fourier $U$-statistics
Yumiharu Nakano
Subjects: Optimization and Control (math.OC); Machine Learning (cs.LG); Numerical Analysis (math.NA); Machine Learning (stat.ML)

We study the subclass of potential mean-field games in which the running interaction cost and the terminal target cost are both expressed through reproducing-kernel maximum mean discrepancy (MMD) penalties, and develop a computational framework that exploits this kernel structure. Both costs are estimated from finite-sample empirical distributions using a random Fourier U-statistic representation that is unbiased and has linear cost in the batch size. The drift of the controlled diffusion is parametrized by a neural network and trained via stochastic gradient descent. For population near-minimizers we prove convergence to the terminal-constrained problem as the penalty diverges, and show that the same limit is recovered almost surely when such controls are evaluated on independent finite samples under explicit coupling conditions on the penalty, random-feature count and sample size. The framework includes the kernel-MMD-penalty Schr{ö}dinger bridge problem as the special case of a vanishing interaction cost. Numerical experiments illustrate the method on the Schr{ö}dinger bridge problem in dimensions up to one hundred, and on an electric vehicle charging coordination problem with per-vehicle physical heterogeneity, where a control-dependent aggregate-demand cost produces a trade-off between peak normalized charging power and the terminal service level.

[54] arXiv:2608.03246 (replaced) [pdf, html, other]
Title: Optimal Oracle Complexity for Weakly Convex Optimization
Jiajin Li, Siyu Pan
Comments: 28 pages, 1 figure. This version extends the lower bound to randomized first-order algorithms and a substantially broader initial-gap regime, with revised proofs and exposition
Subjects: Optimization and Control (math.OC)

We study the oracle complexity of finding $\epsilon$-stationary points of $\rho$-weakly convex and $G$-Lipschitz functions, where stationarity is measured by the gradient norm of the Moreau envelope. We consider an exact first-order oracle that returns both the function value and the full subdifferential at every query point. For sufficiently small $\epsilon>0$, we prove that every first-order algorithm, including randomized algorithms, requires $\Omega({\rho G^2\Delta}/{\epsilon^4})$ oracle queries to succeed with any fixed positive probability, where $f(0)-\inf f\le\Delta$. This lower bound matches the best known deterministic and stochastic first-order upper bounds, up to universal constants, and establishes the optimal randomized and deterministic first-order oracle complexity. The result separates smooth nonconvex and nonsmooth weakly convex optimization: the additional $\epsilon^{-2}$ factor is intrinsic to nonsmooth geometry rather than to stochastic oracle noise.

[55] arXiv:2608.03465 (replaced) [pdf, html, other]
Title: A Spectral Proximal DC Algorithm with Nonmonotone Line Search and Extrapolation
Kelin Wu, Hongpeng Sun
Subjects: Optimization and Control (math.OC); Numerical Analysis (math.NA)

The paper proposes a novel proximal difference-of-convex (DC) algorithmic framework to solve general nonconvex, nonsmooth optimization problems. The basic pBBDCA method combines Barzilai-Borwein (BB) parameter selection with a nonmonotone directional line search, requiring only one proximal solve per iteration. We also develop two extrapolated variants using restart and safeguarding mechanisms. The global convergence of the proposed framework is rigorously established under the Kurdyka-Łojasiewicz property. Numerical experiments on the SCAD-regularized least squares problem and graph Ginzburg-Landau image segmentation models demonstrate that the proposed methods achieve highly competitive efficiency and accuracy.

[56] arXiv:2608.15035 (replaced) [pdf, html, other]
Title: Accelerating Stochastic Gradient Method Using Large Stepsize for Smooth Strongly Convex Minimization
Luwei Bai, Baoyu Zhou
Comments: 46 pages, 5 figures
Subjects: Optimization and Control (math.OC)

We study predetermined stepsize schedules for stochastic gradient methods on m-strongly convex, M-smooth objectives with conditionally unbiased gradient estimates and bounded conditional variance. We derive an exact tight analysis on every schedule with stepsizes in [0,2/(M+1)]. We then characterize the globally optimal one- and two-step schedules over all nonnegative stepsizes. Both optimal schedules depend on the ratio of the initial optimality gap to the noise level and exhibit several distinct regimes. As the influence of stochastic noise diminishes, the optimal stepsizes become larger, reflecting a balance between the benefits of faster iterate convergence and the perturbations induced by stochastic noise. These solutions motivate serial and silver composition rules with explicit performance bounds. For arbitrary horizons, we construct a schedule combining a balanced Silver prefix with constant and decreasing short steps, with the prefix length selected through explicit thresholds. Its certified upper bound never exceeds the exact optimum over all short-step schedules and is strictly smaller whenever a nonempty Silver prefix is selected. Finally, we prove global optimality of the proposed schedule on an explicit sufficient interval of the initial-error-to-variance ratio.

[57] arXiv:2608.23810 (replaced) [pdf, html, other]
Title: Let AEDs Move: Urban Mobility Enhanced Defibrillator Deployment
Bahar D Viniche, Sheng Liu, Nooshin Salari
Subjects: Optimization and Control (math.OC)

Early defibrillation is critical to survival after out-of-hospital cardiac arrest (OHCA). However, public-access defibrillation programs rely largely on stationary automated external defibrillators (AEDs), whose fixed locations cannot adapt to spatial and temporal variation in OHCA demand and which a bystander must retrieve on foot. We study a platform-enabled deployment model in which ride-hailing vehicles carry AEDs that can be dispatched directly to OHCA locations. Holding total AED supply fixed, we evaluate the full range of allocations between stationary and mobile deployment. Our analysis draws on public AED inventories, EMS-reported OHCA incidents, and ride-hailing mobility data from New York City and Toronto. Reallocating part of the supply to vehicles substantially shortens AED response times, but the benefit is non-monotonic in the mobile share. Under a specification common to both cities, mean improvement reaches approximately 2.9 minutes in each, yet the preferred allocation differs: it peaks with 55% of AEDs mobile in Toronto, whereas New York City retains near-maximum gains across 40%-80%, with further variation across boroughs. The preferred allocation also depends on the response-time target, where coverage within 3 and 5 minutes favors greater mobility while 10-minute coverage favors retaining part of the stationary network. Toronto's vehicle-level data further show that targeting AEDs to active vehicles is essential and with such targeting, restricting dispatch to idle vehicles, which disrupts no passenger trip, does not reduce the gain, and allowing any active equipped vehicle to respond raises it to 5.7 minutes under full mobilization. Mobilizing AEDs can therefore materially improve access, and the stationary-mobile mix must be tailored to local coverage, fleet activity, and dispatch rules.

[58] arXiv:2609.15936 (replaced) [pdf, html, other]
Title: An Adaptive Linesearch-free Method for Monotone Variational Inequalities under Local Lipschitz Continuity
Hongjia Ou, Andreas Themelis, Puya Latafat
Subjects: Optimization and Control (math.OC)

The forward-reflected-backward (FRB) splitting solves monotone inclusions involving the sum of a maximally monotone operator and a monotone Lipschitz continuous operator. Each iteration performs one resolvent step and one evaluation of the Lipschitz operator, together with a reflection term with coefficient equal to one that reuses the operator value from the previous iteration. We consider the setting where the maximally monotone operator is the subdifferential of a proper, closed, convex function. We identify the admissible range of constant reflection coefficients, namely all values larger than one half for a sufficiently small stepsize. We then propose a linesearch-free adaptive variant of FRB for locally Lipschitz continuous operators, in which both the stepsize and the reflection coefficient change across iterations. The stepsize is computed in closed form from a simple local Lipschitz estimate, without knowledge of a global Lipschitz constant, while keeping the iteration cost of FRB. The analysis of both methods relies on a new Lyapunov function that combines the distance to a solution, a residual, and a gap-related term. Alongside nonasymptotic rates, we derive explicit lower bounds on the generated stepsizes, and report numerical experiments on minimax problems, equilibrium models, and regularized regression.

[59] arXiv:2610.37444 (replaced) [pdf, html, other]
Title: DREAM: Delay-Resistant Acceleration with Momentum
Mert Gürbüzbalaban, M. Hamza Rafi, Necdet Serhat Aybat, Waheed U. Bajwa
Subjects: Optimization and Control (math.OC)

With exact, delay-free gradients, Nesterov's accelerated gradient (NAG) achieves optimal $\sqrt{\kappa}$ dependence for $L$-smooth, $\mu$-strongly convex objectives, where $\kappa=L/\mu$; heavy-ball (HB) achieves the same scaling on strongly convex quadratics. In asynchronous and distributed settings, gradients can be inexact, with delays varying across workers and iterations. We isolate staleness using a fixed delay $\tau\ge1$, replacing $\nabla f(y_k)$ with $\nabla f(y_{k-\tau})$ at iteration $k$. Arjevani, Shamir, and Srebro establish a worst-case lower bound of $\Omega(\tau\sqrt{\kappa}\log(1/\varepsilon))$ iterations to reach objective error at most $\varepsilon$ as $\varepsilon\to0$. We show that for any fixed positive delay, neither NAG nor HB retains $\sqrt{\kappa}$-acceleration on strongly convex quadratics when updating every iteration, for any stable constant step size and positive momentum. We introduce delay-resistant accelerated momentum (DREAM), which decouples gradient-query extrapolation from momentum and uses explicit delay-dependent parameters. DREAM attains a linear convergence rate $1-\Theta(1/(\tau\sqrt{\kappa}))$ for all smooth strongly convex objectives as $\kappa\to\infty$, yielding complexity matching the delayed first-order lower bound up to logarithmic factors. Our proof combines characteristic-polynomial stability analysis on quadratics with an analytic Zames--Falb integral quadratic constraint certificate. We extend DREAM to smooth convex objectives, obtaining near-optimal iteration complexity up to logarithmic factors for sufficiently small target accuracy. Experiments on strongly convex quadratics and regularized logistic regression corroborate DREAM's accelerated condition-number dependence under delay, unlike NAG and HB, and suggest this scaling persists when parameters use a conservative bound on the true delay.

[60] arXiv:2611.10286 (replaced) [pdf, html, other]
Title: Costate-Conditioned Dynamic Programming for Parallel-in-Time Solution of Equality-Constrained Optimal Control Problems
Tom Lefebvre
Subjects: Optimization and Control (math.OC)

Gradient based trajectory optimisation algorithms spend most of their per-iteration time on a structured Newton--KKT system. The system corresponds to an equality-constrained linear-quadratic optimal control problem and dedicated Riccati-type solvers exhibit linear complexity in the horizon length. Parallel-in-time algorithms exhibit the potential to reduce the complexity to logarithmic in the horizon by distributing several subtrajectories over multiple cores and then stitch together the results by means of an associative scan. Existing exact formulations condition each subtrajectory on its terminal state. Each subproblem then carries the reachability to that terminal state explicitly as constraint rows. The associated reachability decision however is ill-conditioned for weakly actuated systems. In this work we condition each subtrajectory on the terminal costate instead. The resulting subproblems can now be computed by a free-end Riccati recursion without reachability carrying rows. This approach effectively compensates for the ill-conditioning of weakly actuated systems but requires additional bookkeeping with Lagrangian multipliers when stitching together the various subproblems. Numerically the method reproduces a direct solution of the primal-dual Newton-KKT system to $10^{-12}$ in relative terms, on a weakly controllable chain of masses. The accurracy rivals that of a sequential recursion, whereas terminal-state conditioning loses up to six digits or fails. Operation counts on a problem with 64 subtrajectories and a horizon of 4096 stages gives a speed-up of 26 compared with a sequential solve.

[61] arXiv:2411.15121 (replaced) [pdf, html, other]
Title: Information geometric regularization of unidimensional pressureless Euler equations yields global strong solutions
Ruijia Cao, Florian Schäfer
Subjects: Analysis of PDEs (math.AP); Differential Geometry (math.DG); Optimization and Control (math.OC)

Partial differential equations describing compressible fluids are prone to the formation of shock singularities, arising from faster upstream fluid particles catching up to slower, downstream ones. In geometric terms, this causes the deformation map to leave the manifold of diffeomorphisms. Information geometric regularization addresses this issue by changing the manifold geometry to make it geodesically complete. Empirical evidence suggests that this results in smooth solutions without adding artificial viscosity. This work makes a first step towards understanding this phenomenon rigorously, in the setting of the unidimensional pressureless Euler equations. It shows that their information geometric regularization has smooth global solutions. By establishing $\Gamma$-convergence of its variational description, it proves convergence of these solutions to entropy solutions of the nominal problem, in the limit of vanishing regularization parameter. A consequence of these results is that manifolds of unidimensional diffeomorphisms with information geometric regularization are geodesically complete.

[62] arXiv:2510.05241 (replaced) [pdf, html, other]
Title: Simultaneous Learning and Optimization via Misspecified Saddle Point Problems
Mohammad Mahdi Ahmadi, Erfan Yazdandoost Hamedani
Comments: Accepted for publication in the Journal of Computational Optimization and Algorithms
Subjects: Machine Learning (cs.LG); Optimization and Control (math.OC)

We study a class of misspecified saddle point (SP) problems, where the optimization objective depends on an unknown parameter that must be learned concurrently from data. Unlike existing studies that assume parameters are fully known or pre-estimated, our framework integrates optimization and learning into a unified formulation, enabling a more flexible problem class. To address this setting, we propose two algorithms based on the accelerated primal-dual (APD) by Hamedani & Aybat 2021. In particular, we first analyze the naive extension of the APD method by directly substituting the evolving parameter estimates into the primal-dual updates; then, we design a new learning-aware variant of the APD method that explicitly accounts for parameter dynamics by adjusting the momentum updates. Both methods achieve a provable convergence rate of $\mathcal{O}(\log K / K)$, while the learning-aware approach attains a tighter $\mathcal{O}(1)$ constant and further benefits from an adaptive step-size selection enabled by a backtracking strategy. Furthermore, we extend the framework to problems where the learning problem admits multiple optimal solutions, showing that our modified algorithm for a structured setting achieves an $\mathcal{O}(1/\sqrt{K})$ rate. To demonstrate practical impact, we evaluate our methods on a misspecified portfolio optimization problem and show superior empirical performance compared to state-of-the-art algorithms.

[63] arXiv:2604.26741 (replaced) [pdf, html, other]
Title: Analytically Characterized Optimal Power Control for Signal-Level-Integrated Sensing, Computing and Communication in Federated Learning
Paul Zheng, Yao Zhu, Xiaopeng Yuan, Yulin Hu, Anke Schmeink
Journal-ref: IEEE Transactions on Wireless Communications, vol. 25, pp. 25941-25958, 2026
Subjects: Information Theory (cs.IT); Signal Processing (eess.SP); Optimization and Control (math.OC)

In next-generation wireless edge systems, efficient functionality integration is essential to address the growing demands of communication, computation, and sensing. Signal-level integrated sensing, computing, and communication (Sig-ISCC) is envisioned, where a single waveform simultaneously supports sensing, computing and communication via over-the-air computation (AirComp). Meanwhile, federated learning (FL) is widely regarded as a promising distributed machine learning framework that enables network intelligence in a privacy-preserving and secure manner, and exhibits strong synergy with AirComp, which alleviates the communication bottleneck of FL. In this paper, we study uplink Sig-ISCC design for AirComp-FL with joint target detection. We formulate the joint power and receive-scaling control problem, where edge devices' transmitted signals should serve both sensing and AirComp purposes. The goal is to minimize the AirComp aggregation distortion subject to a joint target-detection requirement. Although the resulting problem is non-convex in the original variables, we show that it admits an equivalent convex reformulation after a suitable variable transformation. By exploiting analytical optimality properties, we develop a robust, optimal, and polynomial-time-complexity algorithm that efficiently achieves the optimal transmit powers and receive scaling factor. Simulation results validate the optimality and numerical robustness of the proposed algorithm and show its superior FL performance compared to baseline methods.

[64] arXiv:2605.25610 (replaced) [pdf, html, other]
Title: Match classification in the last round of four-team round-robin tournaments
László Csató, András Gyimesi
Comments: 24 pages, 4 figures, 6 tables
Subjects: Physics and Society (physics.soc-ph); General Economics (econ.GN); Optimization and Control (math.OC); Applications (stat.AP)

Classification of matches played in the last rounds of sports competitions is a well-established tool for evaluating tournament designs. Our paper offers the first comparison of deterministic and probabilistic match classification models by analysing the most prominent example of four-team round-robin competitions, the group stage of the FIFA World Cup. We show that both models are highly relevant in practice: all (four) deterministic and (six) probabilistic match types occurred in the 2014 and 2018 FIFA World Cups, respectively. The probabilistic model, which accounts for the relative benefits of attacking and defending, provides a deeper insight; for instance, the competitive matches from the deterministic approach can be of any of the six probabilistic types. Therefore, the probabilistic model is used to decompose the impact of three reforms introduced in the 2026 FIFA World Cup: the expansion to 48 teams, the modified qualification rules, and head-to-head tie-breakers.

[65] arXiv:2608.15444 (replaced) [pdf, html, other]
Title: Eigenvalues of locally positive semidefinite matrices: Non-convexity and Geometry
Jose Acevedo, Grigoriy Blekherman, Sebastian Debus, Seokbin Lee, Cordian Riener
Comments: New section for general non-convexity
Subjects: Rings and Algebras (math.RA); Algebraic Geometry (math.AG); Optimization and Control (math.OC)

A real symmetric matrix is called $d$-locally positive semidefinite if all of its $d \times d$ principal submatrices are positive semidefinite. We investigate the spectral geometry of $d$-locally positive semidefinite matrices. The set of vectors of eigenvalues of $d$-locally positive semidefinite matrices of size $n \times n$ is fully understood and known to be convex when $d \in \{1,n-1,n\}$ \cite{blekherman2022hyperbolic}. In the smallest remaining case $n=4, d=2$, non-convexity of the set of vectors of eigenvalues was proved in \cite{kozhasov2023eigenvalues}, but even in this case the full description was unknown. We provide a basic semialgebraic description of the set of vectors of eigenvalues for $n=4, d=2$ by establishing a Fischer-type inequality for $2$-locally positive semidefinite matrices of size $4 \times 4$. We prove non-convexity for all $n \geq 4$ and $2 \leq d \leq n - 2$. For $d=2$ we prove a natural linear inequality on a linear slice of eigenvalues of locally positive semidefinite matrices which is tight at exactly two points. This gives a stronger proof of non-convexity in this case. For $d \in \{2, n - 2\}$, analyzing this slice involves solving certain non-smooth and non-convex min-max point configuration problems in the complex plane, which could be of independent interest. Similar problems were considered in \cite{nesterenko2024submatrices,sengupta2026submatrices} in the context of matrix decomposition and approximation.

[66] arXiv:2609.06351 (replaced) [pdf, html, other]
Title: A Theory of Information Architecture for Networked Decisions: Freshness, Locality, and Coordination
Scott Moeller, Bhaskar Krishnamachari
Comments: 32 pages, 6 figures. Manuscript source and complete reproduction materials are available at this https URL
Subjects: Systems and Control (eess.SY); Networking and Internet Architecture (cs.NI); Optimization and Control (math.OC)

A broader view of a network can help agents coordinate, but collecting that view takes time. When is a fresh local observation more useful than an older view of the whole network, and when should an agent look only partway across it? We study agents that repeatedly minimize a shared quadratic cost in an environment that evolves independently of their actions. An information architecture specifies what each agent observes and how old those observations are. Each agent uses a synchronized neighborhood snapshot, so extending its scope also ages the nearby readings used in the decision. We assume that the optimal action with complete information is affine in the state, and that the conditional mean of the current state equals the old state scaled by a common exponential decay factor. Using the classical Marschak--Radner decomposition, we derive the delay threshold beyond which fresh local information outperforms a complete but older global view. This comparison allows non-Gaussian states and nonlinear decision rules. We then determine how far an agent should look by balancing the benefit of additional observations against their greater age. In a Gaussian spatial model on a line, we derive the optimal observation radius: looking beyond the local node pays only when information travels more than twice the spatial correlation length in one coherence time. The preferred information architecture also depends on the decision being made. On a fixed 64-node ring, increasing only the reach of the decision weights from 0.7 to 6 hops shifts the optimal observation radius from one to five hops.

[67] arXiv:2611.17143 (replaced) [pdf, other]
Title: A Convex-Analytic Reformulation for Poisoning Attacks on Logistic Regression
Carina Moreira Costa, Maria Eduarda Pinheiro, Martin Schmidt
Subjects: Machine Learning (cs.LG); Optimization and Control (math.OC)

We study data poisoning attacks on logistic regression with $\ell_2$-regularization, where an attacker chooses the features of a fixed number of injected observations with prescribed labels to maximize the logistic loss on the clean training data. Strong convexity of the defender's problem yields an exact reformulation through its first-order optimality conditions. Using convex conjugacy of the softplus function, we replace the sigmoid relations by a single convex optimality-gap inequality and express the attacker's objective as a concave function. A change of variables makes the stationarity condition affine, leaving bounded bilinear equalities as the only nonconvex constraints. We derive finite bounds on all variables involved in these equalities, including a global bound on the coefficient norm and tighter bounds obtained by preprocessing based on convex optimization. Experiments on eight binary classification datasets evaluate attack effectiveness and compare the two reformulations when solved with SCIP. Attacks obtained by local optimization degrade predictive performance on most datasets, while the convex-analytic formulation produces substantially stronger reported dual bounds within a fixed time limit.

Total of 67 entries
Showing up to 2000 entries per page: fewer | more | all
We gratefully acknowledge support from our major funders, member institutions, , and all contributors.
About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab)