Induction and Exploitation of Subgoal Automata for Reinforcement Learning.
Daniel Furelos-BlancoMark LawAnders JonssonKrysia BrodaAlessandra RussoPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- machine learning
- cellular automata
- function approximation
- state space
- inductive learning
- goal oriented
- partial order
- model free
- tree automata
- optimal control
- transfer learning
- dynamic programming
- robotic control
- exploration exploitation tradeoff
- markov decision processes
- rule induction
- inductive inference
- sequence prediction
- program synthesis
- probabilistic automata
- finite automata
- reinforcement learning algorithms
- finite state machines
- multi agent
- temporal difference
- action selection
- concept formation
- concept learning
- autonomous learning
- policy search
- optimal policy
- supervised learning
- lattice gas
- learning algorithm