mlsanityData to 14 Oct 2026

Relative Entropy Regularized Reinforcement Learning for Efficient Encrypted Policy Synthesis

arXiv 2506.12358 ↗First posted 14 Jun 2025cs.LG cs.SY eess.SYcited by 0

On the map: Deep reinforcement learning methods → RL for energy, networks and operations

Abstract (arXiv)

We propose an efficient encrypted policy synthesis to develop privacy-preserving model-based reinforcement learning. We first demonstrate that the relative-entropy-regularized reinforcement learning framework offers a computationally convenient linear and ``min-free'' structure for value iteration, enabling a direct and efficient integration of fully homomorphic encryption with bootstrapping into policy synthesis. Convergence and error bounds are analyzed as encrypted policy synthesis propagates errors under the presence of encryption-induced errors including quantization and bootstrapping. Theoretical analysis is validated by numerical simulations. Results demonstrate the effectiveness of the RERL framework in integrating FHE for encrypted policy synthesis.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RL for energy, networks and operations, one of 15 directions in Deep reinforcement learning methods. Builds on 9 of its 27 references that are in the map. First posted on arXiv 14 Jun 2025.

Builds on

9 of 27 references.

References · 9

  1. 2018Cloud-based Quadratic Optimization with Partially Homomorphic Encryptioncited by 4
  2. 2020SARSA(0) Reinforcement Learning over Fully Homomorphic Encryptioncited by 3
  3. 2021Encrypted Value Iteration and Temporal Difference Learning over Leveled Homomorphic Encryptioncited by 2
  4. 2018Cloud-based MPC with Encrypted Datacited by 1
  5. 2019Dynamic controller that operates over homomorphically encrypted data for infinite time horizoncited by 1
  6. 2021Input-Output History Feedback Controller for Encrypted Control with Leveled Fully Homomorphic Encryptioncited by 1
  7. 2021Encrypted dynamic control with unlimited operating time via FIR filterscited by 1
  8. 2022Encrypted extremum seeking for privacy-preserving PID tuning as-a-Servicecited by 1
  9. 2024Ring-LWE based encrypted controller with unlimited number of recursive multiplications and effect of error growthcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Efficient Implementation of Reinforcement Learning over Homomorphic Encryption
  2. 2025On the Sample Complexity of Differentially Private Policy Optimization
  3. 2026Privacy Preserving Reinforcement Learning with One-Sided Feedback
  4. 2024Near-Optimal Reinforcement Learning with Shuffle Differential Privacy
  5. 2026Privacy-Preserving Reinforcement Learning from Human Feedback via Decoupled Reward Modeling
  6. 2019Privacy-preserving Q-Learning with Functional Noise in Continuous State Spaces
  7. 2026CHEHAB RL: Learning to Optimize Fully Homomorphic Encryption Computations
  8. 2025Differentially Private Policy Gradient