Bridging the gap between constant step size stochastic gradient descent and Markov chains

2020/06/01 by Aymeric Dieuleveut, Alain Durmus, Francis Bach · 12 citations

paper · doi:10.1214/19-aos1850

Cited by

• A Defense of the Quadratic Model
• Robust equilibria in continuous games: From strategic to dynamic robustness
• Multi-agent learning under uncertainty: Recurrence vs. concentration
• Large-scale empirical tuning and comparison of default optimizers for variational inference
• Statistical Inference for Linear Functionals of Online Least-squares SGD when t \gtrsim d^1+δ
• Statistical Guarantees for High-Dimensional Stochastic Gradient Descent
• A Stochastic Differential Equation Framework for Multi-Objective LLM Interactions: Dynamical Systems Analysis with Code Generation Applications
• SALAD: Self-Adaptive Link Adaptation
• Accelerated Gradient Methods with Biased Gradient Estimates: Risk Sensitivity, High-Probability Guarantees, and Large Deviation Bounds
• A Unifying Framework for Global Optimization: From Theory to Formalization
• Explainable Learning Rate Regimes for Stochastic Optimization
• High-Order Error Bounds for Markovian LSA with Richardson-Romberg Extrapolation