Dependable Reinforcement Learning via Timed Differential Dynamic Logic.
Runhao WangYuhong ZhangHaiying SunJing LiuPublished in: ISCC (2021)
Keyphrases
- dynamic logic
- reinforcement learning
- reasoning about actions
- modal logic
- function approximation
- imperative programs
- petri net
- state space
- agent programming
- reinforcement learning algorithms
- multi agent
- learning algorithm
- classical first order logic
- optimal policy
- machine learning
- model free
- markov decision processes
- timed automata
- propositional dynamic logic
- optimal control
- fault tolerant
- action selection
- discrete event
- first order logic