Fixed-Horizon Temporal Difference Methods for Stable Reinforcement Learning.
Kristopher De AsisAlan ChanSilviu PitisRichard S. SuttonDaniel GravesPublished in: CoRR (2019)
Keyphrases
- temporal difference methods
- reinforcement learning
- temporal difference
- function approximation
- policy search
- function approximators
- evolutionary methods
- policy evaluation
- reinforcement learning problems
- reinforcement learning algorithms
- td learning
- least squares
- evolutionary algorithm
- evolutionary computation
- model free
- step size
- machine learning
- cost function
- td methods
- action selection
- neural network
- policy iteration
- markov decision processes
- transfer learning
- particle swarm optimization
- state space
- learning process