Graph augmented Deep Reinforcement Learning in the GameRLand3D environment.
Edward BeechingMaxim PeterPhilippe MarcotteJilles DibangoyeOlivier SimoninJoshua RomoffChristian WolfPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- dynamic environments
- multi agent
- complex environments
- random walk
- real time
- connected components
- optimal policy
- graph model
- directed acyclic graph
- optimal control
- graph theory
- multi agent environments
- state space
- mobile robot
- learning process
- function approximation
- graph theoretic
- graph representation
- robocup soccer
- weighted graph
- graph structure
- bipartite graph
- virtual world
- data sets