Enhanced Generalization Through Prioritization and Diversity in Self-Imitation Reinforcement Learning Over Procedural Environments with Sparse Rewards.
Alain AndresDaochen ZhaJavier Del SerPublished in: SSCI (2023)
Keyphrases
- reinforcement learning
- function approximation
- multi agent environments
- state space
- markov decision processes
- reinforcement learning algorithms
- learning algorithm
- high dimensional
- dynamic environments
- temporal difference
- machine learning
- imitation learning
- model free
- real world
- object oriented
- sparse representation
- reward function
- optimal policy
- reward shaping
- learning process
- reinforcement learning methods
- policy iteration
- sparse coding
- action selection
- complex environments
- optimal control
- learning problems
- robotic systems
- learning classifier systems
- control policy
- autonomous robots