Imitation and Reinforcement Learning in Agents with Heterogeneous Actions.
Bob PriceCraig BoutilierPublished in: Canadian Conference on AI (2001)
Keyphrases
- reinforcement learning
- state action space
- action selection
- agent behavior
- learning agent
- function approximation
- multiagent reinforcement learning
- multi agent
- multiple agents
- agent receives
- agent learns
- perceptual aliasing
- learned knowledge
- single agent
- action space
- partial observability
- initially unknown
- multi agent systems
- multiagent systems
- reinforcement learning algorithms
- learning agents
- cooperative
- learning capabilities
- explicitly represent
- partially observable
- mirror neurons
- temporal difference
- optimal policy
- learning algorithm
- decision theoretic
- state space
- multi agent reinforcement learning
- reinforcement learning agents
- reward signal
- model free
- multiagent learning
- joint action
- multi agent environments
- actor critic
- reward function
- dynamic environments
- cognitive agents
- state and action spaces
- software agents
- imitation learning
- robocup soccer
- state action
- intelligent agents
- markov decision processes
- function approximators
- reinforcement learning methods
- stochastic games
- interacting agents
- partially observable markov decision processes
- machine learning
- biologically inspired
- agent model