Two-timescale simultaneous perturbation stochastic approximation using deterministic perturbation sequences

DOI10.1145/858481.858486zbMath1390.65045OpenAlexW2125852847MaRDI QIDQ4564833

Shalabh Bhatnagar, Michael C. Fu, Steven I. Marcus, I-Jeng Wang

Publication date: 12 June 2018

Published in: ACM Transactions on Modeling and Computer Simulation (Search for Journal in Brave)

Full work available at URL: http://eprints.iisc.ac.in/266/1/p180-bhatnagar.pdf

zbMATH Keywords

stochastic approximationHadamard matricessimulation optimizationSPSAtwo-timescale algorithmsdeterministic perturbations

Mathematics Subject Classification ID

Probabilistic models, generic numerical methods in probability and statistics (65C20) Numerical optimization and variational techniques (65K10)

Related Items (10)

Multiscale Q-learning with linear function approximation ⋮ Actor-critic algorithms for hierarchical Markov decision processes ⋮ An adaptive optimization scheme with satisfactory transient performance ⋮ A stability criterion for two timescale stochastic approximation schemes ⋮ Variance-constrained actor-critic algorithms for discounted and average reward MDPs ⋮ Weak convergence of dynamical systems in two timescales ⋮ New algorithms of the Q-learning type ⋮ Simultaneous perturbation Newton algorithms for simulation optimization ⋮ Parallel Simultaneous Perturbation Optimization ⋮ Revisiting the ODE method for recursive algorithms: fast convergence using quasi stochastic approximation

This page was built for publication: Two-timescale simultaneous perturbation stochastic approximation using deterministic perturbation sequences