On Reward-Free RL with Kernel and Neural Function Approximations: Single-Agent MDP and Markov Game.
Shuang QiuJieping YeZhaoran WangZhuoran YangPublished in: CoRR (2021)
Keyphrases
- single agent
- reinforcement learning
- policy gradient
- stochastic games
- multi agent
- state action
- markov decision processes
- multiple agents
- action space
- average reward
- reward function
- learning agent
- optimal policy
- function approximators
- markov decision process
- function approximation
- multi agent systems
- decision problems
- approximation methods
- two player games
- state space
- reinforcement learning algorithms
- multi agent coordination
- exploration strategy
- dynamic programming
- window search
- dynamic environments
- policy evaluation
- partially observable markov decision processes
- long run
- dec pomdps
- learning agents
- partially observable
- markov decision problems
- discounted reward
- total reward
- state and action spaces
- continuous state
- model free
- game theory
- inverse reinforcement learning
- finite state
- planning under uncertainty
- reinforcement learning methods
- nash equilibria
- evaluation function
- sufficient conditions
- decision makers
- decision making
- control policy
- learning algorithm