A Cooperation Graph Approach for Multiagent Sparse Reward Reinforcement Learning.
Qingxu FuTenghai QiuZhiqiang PuJianqiang YiWanmai YuanPublished in: IJCNN (2022)
Keyphrases
- reinforcement learning
- multi agent
- cooperative
- multi agent systems
- graph representation
- function approximation
- random walk
- multiagent systems
- multiple agents
- multiagent learning
- single agent
- graph theory
- intelligent agents
- markov decision processes
- reinforcement learning algorithms
- learning agents
- temporal difference
- high dimensional
- dynamic programming
- state space
- autonomous agents
- graph structure
- structured data
- eligibility traces
- action selection
- model free
- graph model
- supervised learning
- optimal policy
- directed graph
- machine learning
- densely connected
- partially observable environments
- average reward
- gaussian graphical models
- reward function
- learning agent
- control policy
- partially observable
- policy gradient
- graph databases
- graph mining
- directed acyclic graph
- signal recovery
- weighted graph
- robocup soccer
- connected components