Augmented Reinforcement Learning for Interaction with Non-expert Humans in Agent Domains.
Mohan SridharanPublished in: ICMLA (1) (2011)
Keyphrases
- reinforcement learning
- human agent
- multi agent
- artificial agents
- agent learns
- action selection
- interacting agents
- cognitive agents
- agent environment
- human users
- human communication
- multi agent systems
- human experts
- intelligent agents
- transfer learning
- conversational agent
- complex domains
- learning agent
- multiagent systems
- agent systems
- autonomous learning
- markov decision processes
- virtual agents
- human agent interaction
- learning capabilities
- deterministic domains
- markov decision process
- partially observable domains
- human players
- learning algorithm
- human computer interaction
- inverse reinforcement learning
- autonomous agents
- multi agent environments
- human operators
- agent negotiation
- multiple agents
- state abstraction
- agent model
- agent technology
- action space
- agent receives
- state action
- software agents
- human human interaction
- real world
- sensory inputs
- state space
- dynamic environments
- optimal policy
- reward function
- robocup soccer
- agent architecture
- function approximation
- mobile agents
- reward shaping
- single agent
- agent communication
- reinforcement learning methods
- learning agents
- agent interactions
- pedagogical agents
- decision theoretic
- human behavior
- domain experts
- machine learning