Enhanced Meta Reinforcement Learning via Demonstrations in Sparse Reward Environments.
Desik RengarajanSapana ChaudharyJaewon KimDileep KalathilSrinivas ShakkottaiPublished in: NeurIPS (2022)
Keyphrases
- reinforcement learning
- state space
- reinforcement learning algorithms
- function approximation
- markov decision processes
- reward function
- temporal difference
- multi agent environments
- machine learning
- model free
- meta level
- optimal policy
- real world
- learning agent
- sparse data
- transfer learning
- learning process
- learning algorithm
- multi agent
- compressive sensing
- partially observable environments
- eligibility traces
- data sets
- reinforcement learning methods
- state action
- average reward
- policy iteration
- compressed sensing
- multi agent systems
- dynamic programming
- supervised learning
- dynamic environments
- finite state
- sparse representation
- learning problems