Sample-Efficient Reinforcement Learning for POMDPs with Linear Function Approximations.
Qi CaiZhuoran YangZhaoran WangPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- function approximators
- function approximation
- dynamic programming
- state space
- markov decision processes
- machine learning
- learning algorithm
- multi agent
- policy search
- partially observable
- policy gradient
- linear functions
- markov decision problems
- approximation methods
- model free
- piecewise linear
- closed form
- optimal policy
- linear approximation
- neural network
- semi infinite programming
- control policy
- partially observable markov decision processes
- reinforcement learning algorithms
- transfer function
- decision problems
- learning process