Encrypted Value Iteration and Temporal Difference Learning over Leveled Homomorphic Encryption.
Jihoon SuhTakashi TanakaPublished in: ACC (2021)
Keyphrases
- temporal difference learning
- homomorphic encryption
- encryption scheme
- markov decision process
- policy iteration
- markov decision processes
- encrypted data
- privacy preserving
- data sharing
- reinforcement learning
- state space
- reinforcement learning algorithms
- third party
- fixed point
- optimal policy
- standard model
- encryption algorithm
- security analysis
- function approximation
- public key
- temporal difference
- sql queries
- ciphertext
- infinite horizon
- evaluation function
- model free
- game playing
- sensitive data
- data storage
- finite state
- initial state
- data hiding
- secret key
- dynamic programming
- service providers
- read write
- data access
- least squares
- partially observable markov decision processes
- data privacy
- transition probabilities
- hash functions
- convergence rate
- neural network