Formalising the Foundations of Discrete Reinforcement Learning in Isabelle/HOL.
Mark ChevallierJacques D. FleuriotPublished in: CoRR (2021)
Keyphrases
- theorem prover
- reinforcement learning
- natural deduction
- theorem proving
- continuous state and action spaces
- first order logic
- higher order logic
- learning algorithm
- sequence prediction
- continuous state
- artificial intelligence
- inference rules
- function approximation
- supervised learning
- temporal difference
- reinforcement learning algorithms
- continuous domains
- model free
- discrete geometry
- policy search
- sequent calculus
- learning process
- formal methods
- continuous state spaces
- optimal policy
- evolutionary algorithm