Asymptotic bias of stochastic gradient search
From MaRDI portal
Publication:1704136
Abstract: The asymptotic behavior of the stochastic gradient algorithm with a biased gradient estimator is analyzed. Relying on arguments based on the dynamic system theory (chain-recurrence) and the differential geometry (Yomdin theorem and Lojasiewicz inequality), tight bounds on the asymptotic bias of the iterates generated by such an algorithm are derived. The obtained results hold under mild conditions and cover a broad class of high-dimensional nonlinear algorithms. Using these results, the asymptotic properties of the policy-gradient (reinforcement) learning and adaptive population Monte Carlo sampling are studied. Relying on the same results, the asymptotic behavior of the recursive maximum split-likelihood estimation in hidden Markov models is analyzed, too.
Recommendations
- scientific article; zbMATH DE number 7625177
- Nonasymptotic Bounds for Stochastic Optimization With Biased Noisy Gradient Oracles
- Exploration of the (non-)asymptotic bias and variance of stochastic gradient Langevin dynamics
- Analysis of biased stochastic gradient descent using sequential semidefinite programs
- Asymptotic and finite-sample properties of estimators based on stochastic gradients
- Non-asymptotic guarantees for sampling by stochastic gradient descent
- scientific article; zbMATH DE number 6860839
- Asymptotic analysis via stochastic differential equations of gradient descent algorithms in statistical and computational paradigms
- On the steplength selection in stochastic gradient methods
- Asymptotic optimality in stochastic optimization
Cites work
- scientific article; zbMATH DE number 3855514 (Why is no real title available?)
- scientific article; zbMATH DE number 48727 (Why is no real title available?)
- scientific article; zbMATH DE number 1321699 (Why is no real title available?)
- scientific article; zbMATH DE number 1005357 (Why is no real title available?)
- scientific article; zbMATH DE number 1972910 (Why is no real title available?)
- scientific article; zbMATH DE number 1753152 (Why is no real title available?)
- scientific article; zbMATH DE number 1405930 (Why is no real title available?)
- A Dynamical System Approach to Stochastic Approximations
- Analyticity, Convergence, and Convergence Rate of Recursive Maximum-Likelihood Estimation in Hidden Markov Models
- Applications of a Kushner and Clark lemma to general classes of stochastic algorithms
- Approximate Dynamic Programming
- Chain recurrence, semiflows, and gradients
- Convergence and robustness of the Robbins-Monro algorithm truncated at randomly varying bounds
- Gradient Convergence in Gradient methods with Errors
- Inference in hidden Markov models.
- Introduction to Stochastic Search and Optimization
- Markov Chains and Stochastic Stability
- Nonlinear systems.
- Nonlinear time series. Theory, methods and applications with R examples
- On gradients of functions definable in o-minimal structures
- On semi- and subanalytic geometry
- OnActor-Critic Algorithms
- 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
- Robustness analysis for stochastic approximation algorithms
- Semianalytic and subanalytic sets
- Stochastic Approximations and Differential Inclusions
- Stochastic approximation and its applications
- Stochastic approximation. A dynamical systems viewpoint.
- Sur le problème de la division
- The O.D.E. Method for Convergence of Stochastic Approximation and Reinforcement Learning
- The geometry of critical and near-critical values of differentiable mappings
Cited in
(17)- Particle-based online estimation of tangent filters with application to parameter estimation in nonlinear state-space models
- Maximum likelihood estimation of regularization parameters in high-dimensional inverse problems: an empirical Bayesian approach. II: Theoretical analysis
- Convergence and convergence rate of stochastic gradient search in the case of multiple and non-isolated extrema
- Bayesian Variational Inference for Exponential Random Graph Models
- scientific article; zbMATH DE number 2040718 (Why is no real title available?)
- On maximum a posteriori estimation with Plug \& Play priors and stochastic gradient descent
- Convergence rates for stochastic approximation: biased noise with unbounded variance, and applications
- Bridging the gap between constant step size stochastic gradient descent and Markov chains
- Bias of particle approximations to optimal filter derivative
- Analysis of biased stochastic gradient descent using sequential semidefinite programs
- Convergence rates for optimised adaptive importance samplers
- Discrepancy-based inference for intractable generative models using quasi-Monte Carlo
- Asymptotic and finite-sample properties of estimators based on stochastic gradients
- scientific article; zbMATH DE number 7625177 (Why is no real title available?)
- Stochastic approximation with discontinuous dynamics, differential inclusions, and applications
- Stability of optimal filter higher-order derivatives
- Gradient estimation with simultaneous perturbation and compressive sensing
This page was built for publication: Asymptotic bias of stochastic gradient search
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1704136)
