Reinforcement learning with rare significant events: direct policy search vs. gradient policy search.
Paul EcoffetNicolas FontbonneJean-Baptiste AndréNicolas BredèchePublished in: GECCO Companion (2021)
Keyphrases
- direct policy search
- policy search
- reinforcement learning
- policy gradient
- reinforcement learning algorithms
- mountain car
- reinforcement learning methods
- function approximation
- continuous state
- state space
- dynamic programming
- markov decision processes
- function approximators
- partially observable markov decision processes
- multi agent
- control problems
- machine learning
- temporal difference
- optimal policy
- neural network
- model free
- transfer learning
- supervised learning
- adaptive control
- radial basis function
- learning process