Baseline: practical control variates for agent evaluation in zero-sum domains.
Joshua DavidsonChristopher ArchibaldMichael BowlingPublished in: AAMAS (2013)
Keyphrases
- real world
- multi agent systems
- control system
- action selection
- decision making
- multi agent
- control method
- data sets
- agent architecture
- optimal control
- control strategy
- autonomous agents
- multiagent systems
- dynamic environments
- real time
- software agents
- transfer learning
- evaluation measures
- learning environment
- multiple agents
- single agent
- adjustable autonomy