Pearl: A Production-ready Reinforcement Learning Agent.
Zheqing ZhuRodrigo de Salvo BrazJalaj BhandariDaniel JiangYi WanYonathan EfroniLiyuan WangRuiyang XuHongbo GuoAlex NikulkovDmytro KorenkevychÜrün DoganFrank ChengZheng WuWanqiao XuPublished in: CoRR (2023)
Keyphrases
- learning agent
- reinforcement learning
- state space
- reinforcement learning algorithms
- learning algorithm
- solving problems
- learning tasks
- learning capabilities
- selective perception
- learning process
- function approximation
- single agent
- reward function
- temporal difference
- model free
- dynamic environments
- multi agent
- optimal policy
- dynamic programming
- search algorithm
- mixed initiative
- artificial intelligence