Program Synthesis Guided Reinforcement Learning for Partially Observed Environments.
Yichen YangJeevana Priya InalaOsbert BastaniYewen PuArmando Solar-LezamaMartin C. RinardPublished in: NeurIPS (2021)
Keyphrases
- program synthesis
- partially observed
- reinforcement learning
- recursive programs
- state space
- dynamic environments
- dynamic programming
- model free
- multi agent environments
- real world
- optimal policy
- markov decision processes
- learning problems
- inductive logic programming
- function approximation
- temporal difference
- machine learning
- expert systems
- artificial intelligence
- decision trees
- reinforcement learning algorithms
- supervised learning
- multi agent
- domain knowledge
- learning process