Offline Reinforcement Learning with Implicit Q-Learning.
Ilya KostrikovAshvin NairSergey LevinePublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- function approximation
- reinforcement learning algorithms
- state space
- model free
- real time
- temporal difference learning
- markov decision processes
- multi agent reinforcement learning
- optimal policy
- action selection
- continuous state and action spaces
- learning algorithm
- reinforcement learning methods
- relational reinforcement learning
- stochastic approximation
- markov decision process
- control problems
- machine learning
- state action space
- dynamic programming
- learning classifier systems
- multi agent
- approximate dynamic programming
- rl algorithms
- temporal difference
- learning capabilities
- function approximators
- learning agent
- learning agents
- partially observable
- average reward
- state abstraction
- policy search
- learning process
- eligibility traces
- cooperative