Skill-Based Reinforcement Learning with Intrinsic Reward Matching.
Ademi AdenijiAmber XiePieter AbbeelPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- state space
- matching algorithm
- function approximation
- learning algorithm
- reinforcement learning algorithms
- reward function
- image matching
- markov decision processes
- partially observable environments
- eligibility traces
- model free
- machine learning
- graph matching
- pattern matching
- shape matching
- dynamic programming
- optimal policy
- feature points
- supervised learning
- learning capabilities
- average reward
- total reward
- matching process
- multi armed bandit
- control policy
- partially observable
- action selection
- multi agent
- learning process