Security Analysis of Safe and Seldonian Reinforcement Learning Algorithms.
Pinar OzisikPhilip S. ThomasPublished in: NeurIPS (2020)
Keyphrases
- security analysis
- reinforcement learning algorithms
- reinforcement learning
- state space
- model free
- markov decision processes
- smart card
- temporal difference
- public key
- reinforcement learning problems
- security requirements
- function approximation
- reinforcement learning methods
- learning algorithm
- dynamic environments
- user authentication
- authentication protocol
- optimal policy
- low cost
- encryption scheme
- monte carlo
- generative model
- supervised learning