Mutual Information as Intrinsic Reward of Reinforcement Learning Agents for On-demand Ride Pooling.
Xianjie ZhangJiahao SunChen GongKai WangYifei CaoHao ChenYu LiuPublished in: AAMAS (2024)
Keyphrases
- mutual information
- reinforcement learning agents
- reinforcement learning
- dynamic environments
- information theoretic
- image registration
- feature selection
- function approximation
- similarity measure
- multi agent
- multi agent environments
- state abstraction
- transfer learning
- learning algorithm
- machine learning
- reward function
- long run
- reinforcement learning algorithms
- learning agent
- heuristic search
- pairwise
- data mining