307 papers
This paper establishes theoretical conditions under which Monotone Operator Equilibrium Networks maintain convergence and bounded error under weight quantization by analyzing spectral perturbations, and validates these findings through experiments showing that quantization-aware training can recover provable convergence at four-bit precision.
This paper proposes a noise-tolerant regularized quasi-Newton method with a relaxed Armijo-type line search that achieves a global convergence rate of for smooth nonconvex optimization under inexact function values, demonstrating superior robustness and competitive efficiency in experiments involving both artificial noise and low-precision arithmetic.
This paper presents a Lanczos tau method for approximating and optimizing the -norm of delay differential algebraic systems, proving its convergence and stability under specific conditions while deriving efficient gradient formulas for robust controller synthesis and demonstrating accelerated convergence through spline-based extensions.
This paper proposes distributed dynamical algorithms that enable agents with only local data and no raw data sharing to collectively compute global system certificates, specifically solving the Lyapunov equation for stability verification and the algebraic Riccati equation for optimal LQR control, with guarantees of exact convergence and robustness.
This paper establishes the first time-uniform error estimate for a fully discrete Luenberger observer applied to the one-dimensional barotropic Euler equations using mixed finite elements and implicit Euler time integration, demonstrating convergence that depends on initial errors, discretization parameters, and measurement noise via a modified relative energy technique.
This paper establishes sharp rates of propagation of chaos for McKean-Vlasov equations with non-linear measure-dependent coefficients, applying these results to derive uniform-in-time convergence guarantees for mean field Langevin dynamics, control, and games by combining the BBGKY hierarchy with weak propagation techniques.
This paper solves Merton's portfolio optimization problem in a multivariate fake stationary affine Volterra setting by employing a martingale optimality principle and a Riccati backward stochastic differential equation to derive semi-closed form optimal strategies, overcoming the challenges posed by non-Markovian and non-semimartingale dynamics.
This paper establishes the existence and uniqueness of equilibrium for a stochastic mean-field game of optimal investment involving identical firms interacting through a time-dependent price, covering both finite and infinite time horizons as well as a deterministic counterpart.
This paper introduces a Partitioned Optimization Framework (POf) that reformulates complex optimization problems by identifying variable subsets that render subproblems tractable, and proposes a Derivative-Free Partitioned Optimization method (DFPOm) that efficiently solves these reformulated problems to find global solutions, demonstrating its effectiveness in both infinite-dimensional optimal control and finite-dimensional composite greybox applications.
This paper presents formal proofs in Lean 4 of Farkas-like theorems over linearly ordered fields and extends duality theory to linear optimization problems where coefficients may take infinite values.
This paper introduces a massively parallel heuristic that approximates Graver basis directions by solving nonconvex continuous problems, thereby overcoming the computational bottlenecks of the augmentation scheme and achieving performance comparable to advanced solvers on nonlinear integer optimization instances.
This paper provides the first complete characterization of optimal dynamic portfolio choice for monotone mean-variance utility in asset models with independent returns under minimal assumptions, establishing a link between maximal utility and the monotone Sharpe ratio while deriving conditions under which classical mean-variance efficient portfolios remain optimal.
This paper establishes the existence of equilibria for a class of Mean Field Games involving reflected stochastic differential equations by utilizing the framework of relaxed controls and martingale problems.
This paper introduces a variant of Polyak's stepsizes for entropic mirror descent to solve linear systems without restrictive domain assumptions, establishing sublinear and linear convergence rates, strengthening -norm implicit bias bounds, and generalizing results to arbitrary convex -smooth functions while proposing an exponentiation-free alternative method.
This study introduces a computational framework combining Design-by-Morphing and Bayesian optimization to generate undulatory swimming profiles that achieve 16%–35% higher propulsive efficiency than traditional anguilliform and carangiform modes by optimizing kinematic parameters and redistributing energy through favorable surface stress distributions.
This paper introduces the concept of dependent reachable sets for two-agent pursuit scenarios, characterizing their geometric bounds and shape through theoretical analysis and simulations of the constant bearing pursuit strategy.
This paper proposes randomized stochastic gradient schemes with smoothing techniques to solve nonconvex nonsmooth stochastic potential games under uncertainty, achieving optimal sample complexity and asymptotic convergence without relying on stringent growth conditions or local convexity.
StochasticBarrier.jl is an open-source Julia toolbox that efficiently synthesizes Stochastic Barrier Functions for verifying the safety of discrete-time stochastic systems using Sum-of-Squares and piecewise constant optimization methods, demonstrating superior speed, scalability, and safety bounds compared to state-of-the-art tools across over 30 case studies.
This paper establishes a general mathematical framework for variational problems in quantum thermodynamics with measurement constraints, leveraging non-commutative optimal transport to analyze dual formulations and zero-temperature limits while tailoring the approach to quantum state tomography and developing convergent computational algorithms.
This paper introduces the concept of marking data-informativity and develops corresponding algorithms to design valid nonblocking supervisors for unknown discrete-event systems using available behavioral data, while also addressing cases where data is insufficient through restricted informativity and informatizability frameworks.