$\mathrm{SO}(2)$-Equivariant Reinforcement Learning.
Dian WangRobin WaltersRobert PlattPublished in: ICLR (2022)
Keyphrases
- reinforcement learning
- function approximation
- rotation invariant
- learning algorithm
- state space
- robotic control
- optimal policy
- control problems
- multi agent
- reinforcement learning algorithms
- temporal difference
- markov decision processes
- optimal control
- markov decision process
- action selection
- model free
- dynamic programming
- learning process
- information retrieval
- learning classifier systems
- real time
- transfer learning
- partially observable
- information systems
- function approximators
- temporal difference learning
- policy search
- neural network