Interactive Explanations: Diagnosis and Repair of Reinforcement Learning Based Agent Behaviors.
Christian Arzate CruzTakeo IgarashiPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- multi agent
- action selection
- learning agent
- single agent
- selective perception
- multi agent systems
- multiagent systems
- intelligent agents
- state abstraction
- multi agent environments
- autonomous learning
- agent interactions
- state action
- reward function
- learning capabilities
- function approximation
- model based diagnosis
- agent behavior
- conversational agents
- agent receives
- medical diagnosis
- agent technology
- exploration strategy
- multiple agents
- model free
- fault diagnosis
- embodied conversational agents
- real robot
- reinforcement learning algorithms
- learning agents
- reinforcement learning agents
- reward shaping
- repair actions
- decision making
- state space
- neural network
- autonomous agents
- partially observable
- agent learns
- mobile agents
- multiagent reinforcement learning
- computer graphics
- markov decision processes
- social norms
- human behavior
- unknown environments
- agent architecture
- agent model
- pedagogical agents
- markov decision process
- software agents
- sensory inputs
- state transitions
- optimal policy
- action space
- expert systems
- machine learning