Improving Interactive Reinforcement Agent Planning with Human Demonstration.
Guangliang LiRandy GomezKeisuke NakamuraJinying LinQilei ZhangBo HePublished in: CoRR (2019)
Keyphrases
- stochastic domains
- multi agent systems
- human users
- action selection
- artificial agents
- decision theoretic
- conversational agents
- intelligent agents
- reinforcement learning
- world model
- autonomous agents
- mixed initiative
- planning process
- decision making
- goal directed
- agent receives
- multiagent systems
- reward shaping
- user friendly
- human behavior
- multi agent
- agent teams
- human communication
- conversational agent
- reactive agents
- external events
- semantic web service composition
- virtual agents
- human agent
- plan execution
- single agent
- decision support
- planning problems
- human subjects
- agent technology
- multiple agents
- agent systems
- virtual humans
- agent interactions
- human robot interaction
- human interaction
- goal oriented
- visualization tool
- computational models
- learning agent
- domain independent
- heuristic search
- computer graphics
- action selection mechanism
- state space