Think Too Fast Nor Too Slow: The Computational Trade-off Between Planning And Reinforcement Learning.
Thomas M. MoerlandAnna DeichlerSimone BaldiJoost BroekensCatholijn M. JonkerPublished in: CoRR (2020)
Keyphrases
- trade off
- reinforcement learning
- action selection
- function approximation
- partially observable
- reinforcement learning algorithms
- planning problems
- goal oriented
- temporal difference
- model free
- decision theoretic
- partial observability
- state space
- learning algorithm
- neural network
- deterministic domains
- markov decision problems
- domain independent
- optimal policy
- machine learning