POMO: Policy Optimization with Multiple Optima for Reinforcement Learning.
Yeong-Dae KwonJinho ChooByoungjip KimIljoo YoonSeungjai MinYoungjune GwonPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- optimization algorithm
- optimal policy
- global optimization
- evolutionary algorithm
- optimization problems
- optimal solution
- decision problems
- partially observable
- optimization process
- learning algorithm
- optimization method
- linear programming
- action selection
- multi objective
- action space
- policy search
- partially observable domains