Adaptive Genomic Evolution of Neural Network Topologies (AGENT) for State-to-Action Mapping in Autonomous Agents.
Amir BehjatSharat ChidambaranSouma ChowdhuryPublished in: CoRR (2019)
Keyphrases
- autonomous agents
- network topologies
- multi agent systems
- dynamic environments
- multiagent systems
- multi agent
- incomplete information
- heterogeneous agents
- open systems
- resource bounded
- complex environments
- reward signal
- interacting agents
- action space
- coalition formation
- network topology
- mental states
- bdi agents
- agent negotiation
- state space
- multi agent environments
- rational agents
- learning agents
- intelligent behavior
- markov decision processes
- intelligent agents
- multi issue
- reinforcement learning