Equivariant Q Learning in Spatial Action Spaces.
Dian WangRobin WaltersXupeng ZhuRobert Platt Jr.Published in: CoRR (2021)
Keyphrases
- action space
- state space
- reinforcement learning
- action selection
- continuous state spaces
- state action
- continuous state
- markov decision processes
- reinforcement learning methods
- single agent
- multi agent
- state and action spaces
- cooperative
- optimal policy
- reinforcement learning algorithms
- function approximators
- stochastic processes
- learning algorithm
- function approximation
- real valued
- dynamic programming
- control policies
- skill learning
- policy iteration
- model free
- state variables
- planning problems
- heuristic search
- particle filter