Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Multi-Armed Bandit-Based Participant Selection Method for Federated Recommendation Systems

Jintao Liu, Mohammad Goudarzi, Adel Nadjaran Toosi

Abstract

Federated Recommendation Systems (FRS) enable privacy-preserving model training by keeping user data on edge devices. However, the practical deployment of FRS in Edge-Cloud environments faces significant challenges due to system and statistical heterogeneity. Existing FRS participant selection strategies struggle to dynamically balance the trade-off between model convergence speed and recommendation quality in such volatile environments. To address this, we formulate the FRS participant selection problem as a normalized utility cost addressing the model quality and system efficiency. Next, we propose a dynamic participant selection framework incorporating a Multi-Armed Bandit (MAB)-based solver for multimodal FRS. We design a client-utility function that jointly evaluates historical Client Performance Reputation, data quality, and real-time system latency. By leveraging an Upper Confidence Bound strategy, our framework effectively balances the exploration of under-sampled clients with the exploitation of high-performing ones. We validate the proposed approach on a realistic edge-cloud testbed implementation using a multimodal movie-recommendation task. Experimental results demonstrate that our MAB-driven approach outperforms other baselines across eight different data-skew scenarios. Specifically, it improves training efficiency by 32-50% while improving model quality metrics such as Recall@50 by up to around 5%

A Multi-Armed Bandit-Based Participant Selection Method for Federated Recommendation Systems

Abstract

Federated Recommendation Systems (FRS) enable privacy-preserving model training by keeping user data on edge devices. However, the practical deployment of FRS in Edge-Cloud environments faces significant challenges due to system and statistical heterogeneity. Existing FRS participant selection strategies struggle to dynamically balance the trade-off between model convergence speed and recommendation quality in such volatile environments. To address this, we formulate the FRS participant selection problem as a normalized utility cost addressing the model quality and system efficiency. Next, we propose a dynamic participant selection framework incorporating a Multi-Armed Bandit (MAB)-based solver for multimodal FRS. We design a client-utility function that jointly evaluates historical Client Performance Reputation, data quality, and real-time system latency. By leveraging an Upper Confidence Bound strategy, our framework effectively balances the exploration of under-sampled clients with the exploitation of high-performing ones. We validate the proposed approach on a realistic edge-cloud testbed implementation using a multimodal movie-recommendation task. Experimental results demonstrate that our MAB-driven approach outperforms other baselines across eight different data-skew scenarios. Specifically, it improves training efficiency by 32-50% while improving model quality metrics such as Recall@50 by up to around 5%

Paper Structure

This paper contains 29 sections, 23 equations, 4 figures, 4 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Background and Related Work
  3. Multimodal Recommendation System
  4. Federated Learning
  5. Related Work
  6. System Model and Problem Formulation
  7. Multimodal Recommendation Backbone
  8. Modality Projection and Factor-Wise Decomposition
  9. Factor-Wise Multimodal Attention and Prediction
  10. Problem Formulation
  11. Customized Client Performance Reputation
  12. Update Relevance
  13. Data Quality Measure
  14. Aggregate Contribution Score
  15. Per-Client Latency and Normalized Time Cost
  16. ...and 14 more sections

Figures (4)

  • Figure 1: Model Evaluation over Recommendation Metrics
  • Figure 2: Framework of the Upper Confidence Bound-based Multi-Armed Bandit for Federated Recommendation System participant selection.
  • Figure 3: Allocation of data samples to each client under different UBI settings.
  • Figure 4: Area Under the Receiver Operating Characteristic Curve (AUC) over Cumulative Training Time for RPFL, FedAvg, and our method under different data partitioning, with different User Balance Index settings.