311 papers
This paper proposes a novel Hybrid Heuristic-Reinforcement Learning (HHRL) framework that integrates railway-specific heuristics with Q-learning to efficiently solve complex railcar shunting problems involving both one-sided and two-sided classification tracks by decomposing multi-locomotive tasks into manageable subproblems.
This paper proposes a distributionally robust framework for standard quadratic optimization under Wasserstein ambiguity, demonstrating its equivalence to a modified deterministic StQP instance while providing out-of-sample performance guarantees and empirical validation.
This paper establishes the existence of equilibria for non-Markovian mean-field games with unbounded controls and quadratic running costs using a weak formulation approach grounded in new stability results for quadratic-growth generalized McKean-Vlasov BSDEs, thereby removing previous restrictions on model parameter boundedness and time horizons.
This paper presents a specialized lock-free work-stealing queue designed for a master-worker framework in mixed-integer programming solvers that leverages restricted concurrency assumptions to support native bulk operations and achieve constant-latency push performance, significantly outperforming general-purpose implementations like C++ Taskflow in batch processing scenarios.
This paper introduces a novel frequency response formulation for nonlinear systems within the Koopman operator framework, deriving a generalized complex-valued function via Laplace transforms and resolvent theory to enable Bode plot analysis under specific existence conditions.
This paper proposes a nonlinear conjugate gradient method with Wolfe line search for solving unconstrained multiobjective interval optimization problems, providing proofs of global convergence for general and specific algorithmic parameters (Fletcher-Reeves, Conjugate Descent, Dai-Yuan, and modified Dai-Yuan) alongside numerical performance evaluations.
This paper introduces a Bayesian framework for linear programming under learned uncertainty that integrates data-driven posterior updates with robust optimization and scenario-based strategies to provide rigorous, finite-sample feasibility guarantees and improved decision safety compared to naive approaches.
This paper proposes a Newton-based algorithm with an Armijo-like line search strategy to solve multiobjective interval optimization problems by establishing a link between weakly Pareto optimal and Pareto critical points, proving its convergence, and validating its effectiveness through numerical experiments and a portfolio optimization application.
This paper proposes a hierarchical Bayesian dynamic game framework for competitive inventory and pricing under incomplete information, integrating Bayesian learning, strategic belief updating, and a credible-risk criterion to derive a conservative equilibrium that effectively balances profit maximization with uncertainty management.
This paper proposes a policy iteration algorithm for general entropy-regularized time-inconsistent stochastic control problems that converges exponentially to an equilibrium policy by proving the generated value functions form a Cauchy sequence, thereby establishing the global existence and uniqueness of a classical solution to the associated exploratory equilibrium Hamilton–Jacobi–Bellman equation.
This paper establishes robust upper and lower bounds for life insurance functionals under two distinct mortality assumptions—one requiring almost sure consistency and the other allowing for expected deviations with life table constraints—to quantify the impact of fractional age uncertainty on contract valuations without relying on specific parametric models.
This paper establishes a Pontryagin-type stochastic maximum principle for optimal control of McKean-Vlasov stochastic partial differential equations with nonconvex control sets by employing spike variation methods and Lions derivatives within an infinite-dimensional framework.
This paper analyzes the gradient flow dynamics of the value-softmax model, a core component of self-attention, to demonstrate that training inherently drives the system toward low-entropy solutions, thereby providing a theoretical explanation for empirical phenomena like attention sinks and massive activations.
This paper introduces a new variant of the line-based dial-a-ride problem without time windows, proposes a novel MILP formulation and a branch-and-price algorithm based on generating profitable stopping patterns, and demonstrates through computational experiments that both the exact method and a scalable root node heuristic effectively solve large instances with high efficiency and small optimality gaps.
This paper reinterprets NP witness discovery as an information-acquisition process and demonstrates that in a structureless "psocid" model where witnesses are accessed only via equality probes, the exponentially limited information gain per probe fundamentally prevents reliable recovery within polynomial time, thereby revealing an informational origin for exponential search complexity.
This paper introduces ALFCG, the first adaptive, projection-free framework for stochastic composite nonconvex optimization that eliminates the need for global smoothness constants or line search by using self-normalized accumulators to estimate local smoothness, achieving optimal iteration complexity up to logarithmic factors while outperforming state-of-the-art baselines.
This paper presents an optimization-based formulation of the Red Light Green Light (RLGL) algorithm for computing stationary distributions of large Markov chains via Dirichlet-energy minimization and coordinate descent, thereby clarifying its behavior, establishing exponential convergence for specific chain classes, and suggesting practical scheduling strategies to accelerate convergence.
This paper establishes that a notion of higher-order normality, derived from iterated Lie brackets, is sufficient to prevent infimum gaps in impulsive extensions of control-affine systems under an -control topology, thereby extending existing no-gap results beyond the more common -trajectory framework.
This paper establishes and proves the tightness of the optimal bounded stepsize limits for Popov's algorithm in solving generalized monotone variational inequalities, demonstrating a bound of for constrained cases and an improved for unconstrained cases through a novel Lyapunov-type function analysis.
This paper presents experimental verification and numerical simulations demonstrating that the spontaneous symmetry breaking machine (SSBM) can effectively solve combinatorial optimization problems, including the large-scale benchmark, by leveraging its unique principle to explore extremely stable states.