Improving the Efficiency of Off-Policy Reinforcement Learning by Accounting for Past Decisions.
Brett DaleyChristopher AmatoPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- machine learning
- multi agent
- decision making
- genetic algorithm
- database
- computational complexity
- learning process
- optimal policy
- reinforcement learning algorithms
- case study
- image sequences
- bayesian networks
- multiscale
- decision makers
- markov decision processes
- function approximation
- high efficiency
- model free