Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

HERL: Tiered Federated Learning with Adaptive Homomorphic Encryption using Reinforcement Learning

Jiaxang Tang, Zeshan Fayyaz, Mohammad A. Salahuddin, Raouf Boutaba, Zhi-Li Zhang, Ali Anwar

TL;DR

HELP, a Reinforcement Learning-based approach that uses Q-Learning to dynamically optimize encryption parameters, specifically the polynomial modulus degree, $N$, and the coefficient modulus, $q, across different client tiers, is proposed.

Abstract

Federated Learning is a well-researched approach for collaboratively training machine learning models across decentralized data while preserving privacy. However, integrating Homomorphic Encryption to ensure data confidentiality introduces significant computational and communication overheads, particularly in heterogeneous environments where clients have varying computational capacities and security needs. In this paper, we propose HERL, a Reinforcement Learning-based approach that uses Q-Learning to dynamically optimize encryption parameters, specifically the polynomial modulus degree, $N$, and the coefficient modulus, $q$, across different client tiers. Our proposed method involves first profiling and tiering clients according to the chosen clustering approach, followed by dynamically selecting the most suitable encryption parameters using an RL-agent. Experimental results demonstrate that our approach significantly reduces the computational overhead while maintaining utility and a high level of security. Empirical results show that HERL improves utility by 17%, reduces the convergence time by up to 24%, and increases convergence efficiency by up to 30%, with minimal security loss.

HERL: Tiered Federated Learning with Adaptive Homomorphic Encryption using Reinforcement Learning

TL;DR

HELP, a Reinforcement Learning-based approach that uses Q-Learning to dynamically optimize encryption parameters, specifically the polynomial modulus degree, , and the coefficient modulus, $q, across different client tiers, is proposed.

Abstract

Federated Learning is a well-researched approach for collaboratively training machine learning models across decentralized data while preserving privacy. However, integrating Homomorphic Encryption to ensure data confidentiality introduces significant computational and communication overheads, particularly in heterogeneous environments where clients have varying computational capacities and security needs. In this paper, we propose HERL, a Reinforcement Learning-based approach that uses Q-Learning to dynamically optimize encryption parameters, specifically the polynomial modulus degree, , and the coefficient modulus, , across different client tiers. Our proposed method involves first profiling and tiering clients according to the chosen clustering approach, followed by dynamically selecting the most suitable encryption parameters using an RL-agent. Experimental results demonstrate that our approach significantly reduces the computational overhead while maintaining utility and a high level of security. Empirical results show that HERL improves utility by 17%, reduces the convergence time by up to 24%, and increases convergence efficiency by up to 30%, with minimal security loss.
Paper Structure (18 sections, 5 equations, 10 figures, 1 table, 1 algorithm)

This paper contains 18 sections, 5 equations, 10 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Preliminary
  3. Federated Learning
  4. Homomorphic Encryption
  5. Federated Learning with Homomorphic Encryption
  6. Related work
  7. Motivation
  8. Influence of heterogeneity and HE parameter plan.
  9. Adaptive HE parameter plan.
  10. HERL
  11. Overview
  12. Profiling and tiering
  13. RL-based parameter selection
  14. Experiment
  15. Setup
  16. ...and 3 more sections

Figures (10)

  • Figure 1: The influence HE parameter plan on latency and security
  • Figure 2: Impact of different HE parameter plans on latency and precision.
  • Figure 3: Comparing the normalized Security (red) and Accuracy (green), Communication time (orange), and HE time (blue) across six configurations, categorized by client speed (Fast, Moderate, Slow) and HE parameters (High, Low).
  • Figure 4: The relationship between test accuracy and total convergence time across three methods.
  • Figure 5: High level overview of HERL.
  • ...and 5 more figures