Programmatically Interpretable Reinforcement Learning.
Abhinav VermaVijayaraghavan MuraliRishabh SinghPushmeet KohliSwarat ChaudhuriPublished in: CoRR (2018)
Keyphrases
- reinforcement learning
- reinforcement learning algorithms
- state space
- function approximation
- temporal difference learning
- machine learning
- temporal difference
- markov decision processes
- model free
- optimal control
- multi agent
- reinforcement learning methods
- optimal policy
- learning agent
- control problems
- robot control
- direct policy search
- stochastic approximation
- policy search
- relational reinforcement learning
- real time
- action selection
- classification rules
- transfer learning
- sufficient conditions
- supervised learning
- mobile robot
- dynamic programming
- search space
- knowledge base
- information systems
- learning algorithm
- real world