CertRL: Formalizing Convergence Proofs for Value and Policy Iteration in Coq.
Koundinya VajjhaAvraham ShinnarVasily PestunBarry M. TragerNathan FultonPublished in: CoRR (2020)
Keyphrases
- policy iteration
- stochastic approximation
- markov decision processes
- convergence rate
- theorem prover
- model free
- reinforcement learning
- optimal policy
- fixed point
- least squares
- sample path
- policy evaluation
- finite state
- infinite horizon
- markov decision process
- temporal difference
- convergence speed
- linear programming
- monte carlo
- markov decision problems
- first order logic
- average reward
- step size
- optimal control
- stationary points
- linear program
- markov random field
- state space
- dynamic programming
- pairwise
- multi agent
- objective function