POMO: Policy Optimization with Multiple Optima for Reinforcement Learning.
Yeong-Dae KwonJinho ChooByoungjip KimIljoo YoonYoungjune GwonSeungjai MinPublished in: NeurIPS (2020)
Keyphrases
- reinforcement learning
- optimization algorithm
- optimal policy
- evolutionary algorithm
- policy search
- optimization problems
- global optimization
- function approximation
- optimization process
- action selection
- state space
- multi agent
- markov decision process
- machine learning
- optimal control
- partially observable
- policy iteration
- policy evaluation
- approximate dynamic programming