This paper demonstrates that carefully designed adaptive stepsizes enable stochastic model-based optimization methods to achieve an O(1/√K) convergence rate on weakly convex objectives, even without the typical assumption of Lipschitz continuity.
Two-stage stochastic programs with fixed recourse and fixed costs can be optimally solved using a single, carefully chosen "optimal scenario" forecast, potentially outperforming traditional methods that rely on approximating the entire uncertainty distribution.
This paper proposes and analyzes a continuous-time model for stochastic gradient descent with momentum, exploring its convergence properties and proposing a stable discretization scheme for practical application in machine learning optimization.
This paper introduces a novel Regularized Progressive Hedging Algorithm (RPHA) for robust stochastic optimal control, enhancing out-of-sample robustness in energy management systems by incorporating variance penalization and demonstrating superior performance compared to standard methods like Model Predictive Control (MPC) and standard PHA.
This paper introduces ReLU Lagrangian cuts, a new family of nonlinear cuts that provide a tight approximation for solving two-stage stochastic mixed-integer programs (SMIPs) with general mixed-integer decisions in the first stage.
This research paper proves the almost sure convergence of a class of stochastic optimization algorithms, inspired by Hamiltonian dynamics, to stationary points of an objective function under various smoothness and noise conditions, including L-smoothness, (L0, L1)-smoothness, and heavy-tailed noise.
本文提出了一種名為 SMAG 的單迴圈隨機演算法,用於解決一類非光滑、非凸的 DMax 優化問題,並在理論上證明了其具有與當前最佳演算法相同的非漸進收斂速度。
This paper introduces SMAG, a novel single-loop stochastic algorithm designed to efficiently solve a class of non-smooth, non-convex optimization problems, specifically focusing on the difference of weakly convex functions with a max-structure (DMax).
Incorporating a specific type of momentum into the Stochastic Cubic Newton method significantly improves its convergence rate for non-convex optimization problems, enabling convergence for any batch size, including single-sample batches.
This paper introduces novel sign-based stochastic variance reduction algorithms for non-convex optimization, achieving improved convergence rates compared to existing sign-based methods, both in centralized and distributed settings.