Improving Exploration in Evolution Strategies for Deep Reinforcement Learning via a Population of Novelty-Seeking Agents.
Edoardo ContiVashisht MadhavanFelipe Petroski SuchJoel LehmanKenneth O. StanleyJeff ClunePublished in: CoRR (2017)
Keyphrases
- evolution strategy
- reinforcement learning
- action selection
- multi agent
- learning agents
- multi agent systems
- evolutionary algorithm
- multiagent systems
- single agent
- intelligent agents
- differential evolution
- active exploration
- multi agent environments
- agent receives
- multi agent reinforcement learning
- evolutionary programming
- exploration strategy
- entire population
- software agents
- multiple agents
- temporal difference
- learning agent
- function approximation
- numerical optimization
- population dynamics
- autonomous agents
- robocup soccer
- genetic algorithm
- autonomous learning
- optimization methods
- particle swarm optimization algorithm
- multiagent learning
- mutation operator
- cooperative
- markov decision processes
- state space
- reinforcement learning algorithms
- dynamic environments
- optimal policy
- neural network