Asymptotic bias of stochastic gradient search

DOI10.1214/16-AAP1272zbMath1387.49044arXiv1709.00291OpenAlexW2591423585MaRDI QIDQ1704136

Arnaud Doucet, Vladislav B. Tadić

Publication date: 8 March 2018

Published in: The Annals of Applied Probability (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1709.00291

zbMATH Keywords

chain-recurrenceLojasiewicz inequalitiesstochastic gradient searchbiased gradient estimationYomdin theorem

Mathematics Subject Classification ID

Numerical methods based on necessary conditions (49M05) Stochastic programming (90C15) Identification in stochastic control theory (93E12) Stochastic approximation (62L20) Stochastic learning and adaptive control (93E35)

Related Items

Bayesian Variational Inference for Exponential Random Graph Models, Stability of optimal filter higher-order derivatives, Stochastic approximation with discontinuous dynamics, differential inclusions, and applications, Bridging the gap between constant step size stochastic gradient descent and Markov chains, On maximum a posteriori estimation with Plug \& Play priors and stochastic gradient descent, Discrepancy-based inference for intractable generative models using quasi-Monte Carlo, Maximum Likelihood Estimation of Regularization Parameters in High-Dimensional Inverse Problems: An Empirical Bayesian Approach. Part II: Theoretical Analysis, Unnamed Item, Convergence rates for optimised adaptive importance samplers, Particle-based online estimation of tangent filters with application to parameter estimation in nonlinear state-space models, Bias of Particle Approximations to Optimal Filter Derivative

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Particle approximations of the score and observed information matrix in state space models with application to parameter estimation
• Perturbations of set-valued dynamical systems, with applications to game theory
• Stochastic approximation. A dynamical systems viewpoint.
• Convergence and robustness of the Robbins-Monro algorithm truncated at randomly varying bounds
• Semianalytic and subanalytic sets
• On gradients of functions definable in o-minimal structures
• On semi- and subanalytic geometry
• The geometry of critical and near-critical values of differentiable mappings
• Stochastic approximation and its applications
• Chain recurrence, semiflows, and gradients
• Inference in hidden Markov models.
• Applications of a Kushner and Clark lemma to general classes of stochastic algorithms
• Markov Chains and Stochastic Stability
• Robustness analysis for stochastic approximation algorithms
• Introduction to Stochastic Search and Optimization
• OnActor-Critic Algorithms
• Gradient Convergence in Gradient methods with Errors
• A Dynamical System Approach to Stochastic Approximations
• The O.D.E. Method for Convergence of Stochastic Approximation and Reinforcement Learning
• Analyticity, Convergence, and Convergence Rate of Recursive Maximum-Likelihood Estimation in Hidden Markov Models
• Approximate Dynamic Programming
• Stochastic Approximations and Differential Inclusions
• Sur le problème de la division