The actor-critic algorithm as multi-time-scale stochastic approximation.

DOI10.1007/BF02745577zbMath1075.90557OpenAlexW2047364871MaRDI QIDQ5955801

Vijaymohan R. Konda, Vivek S. Borkar

Publication date: 18 February 2002

Published in: Sādhanā (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1007/bf02745577

Mathematics Subject Classification ID

Stochastic approximation (62L20) Markov and semi-Markov decision processes (90C40)

Related Items (1)

Reinforcement learning based algorithms for average cost Markov decision processes

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Nonconvergence to unstable points in urn models and stochastic approximations
• Generalized polynomial approximations in Markovian decision processes
• New method of stochastic approximation type
• Stochastic approximation methods for constrained and unconstrained systems
• Asynchronous stochastic approximation and Q-learning
• Stochastic approximation with two time scales
• \({\mathcal Q}\)-learning
• Feature-based methods for large scale dynamic programming
• Do stochastic algorithms avoid traps?
• Chaotic relaxation
• Estimation and control in discounted stochastic dynamic programming
• A tutorial survey of reinforcement learning

This page was built for publication: The actor-critic algorithm as multi-time-scale stochastic approximation.