Prioritizing Samples in Reinforcement Learning with Reducible Loss.
Shivakanth SujitSomjit NathPedro H. M. BragaSamira Ebrahimi KahouPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- markov decision processes
- state space
- machine learning
- robotic control
- sample points
- data samples
- reinforcement learning algorithms
- model free
- function approximation
- optimal policy
- data sets
- training samples
- transfer learning
- supervised learning
- dynamic programming
- learning algorithm
- policy search
- learning process
- high dimensional
- partially observable
- training data
- function approximators
- temporal difference learning