Improving Reinforcement Learning Performance through a Behavioral Psychology-Inspired Variable Reward Scheme.
Heena RathoreHenry GriffithPublished in: SMARTCOMP (2023)
Keyphrases
- reinforcement learning
- function approximation
- state space
- reinforcement learning algorithms
- multi agent
- computer science
- artificial intelligence
- model free
- classification scheme
- temporal difference
- eligibility traces
- action space
- reward shaping
- human cognition
- partially observable
- learning classifier systems
- computational models
- human computer interaction