Reinforcement learning with temporal logic rewards.
Xiao LiCristian Ioan VasileCalin BeltaPublished in: IROS (2017)
Keyphrases
- temporal logic
- reinforcement learning
- model checking
- markov decision processes
- modal logic
- state space
- reward function
- satisfiability problem
- belief revision
- concurrent systems
- dynamic programming
- temporal knowledge
- reinforcement learning algorithms
- mazurkiewicz traces
- linear temporal logic
- verification method
- learning algorithm
- reward shaping
- multi agent
- optimal policy
- formal specification
- computation tree logic
- partially observable
- heuristic search
- hidden state
- continuous state
- reactive systems
- search space
- multi agent systems
- transition systems
- model checker
- automata theoretic
- model free