Revisiting the Minimalist Approach to Offline Reinforcement Learning.
Denis TarasovVladislav KurenkovAlexander NikulinSergey KolesnikovPublished in: NeurIPS (2023)
Keyphrases
- reinforcement learning
- reinforcement learning algorithms
- state space
- real time
- function approximation
- optimal policy
- markov decision processes
- learning process
- control problems
- model free
- learning algorithm
- temporal difference
- neural network
- policy search
- multi agent reinforcement learning
- temporal difference learning
- action selection
- supervised learning
- dynamic programming
- learning capabilities
- control system
- evolutionary algorithm
- reinforcement learning methods
- stochastic approximation
- continuous state
- multi agent
- computer vision
- partially observable domains