Federated Linear Contextual Bandits

2021/10/27 by Ruiquan Huang, Weiqiang Wu, Huang, Ruiquan +5 · 2 citations
Decision Sciences · Engineering · #Advanced Bandit Algorithms Research #Advanced Wireless Network Optimization #FOS: Computer and information sciences #Information Theory (cs.IT) #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Smart Grid Energy Management

paper · pdf · doi:10.48550/arxiv.2110.14177

openalex publication_date 2021/10/27 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

This paper presents a novel federated linear contextual bandits model, where individual clients face different K-armed stochastic bandits coupled through common global parameters. By leveraging the geometric structure of the linear rewards, a collaborative algorithm called Fed-PE is proposed to cope with the heterogeneity across clients without exchanging local feature vectors or raw data. Fed-PE relies on a novel multi-client G-optimal design, and achieves near-optimal regrets for both disjoint and shared parameter cases with logarithmic communication costs. In addition, a new concept called collinearly-dependent policies is introduced, based on which a tight minimax regret lower bound for the disjoint parameter case is derived. Experiments demonstrate the effectiveness of the proposed algorithms on both synthetic and real-world datasets.

Citations

• Thompson Sampling for Contextual Bandits with Linear Payoffs
• Approximate D-optimal Experimental Design with Simultaneous Size and\n Cost Constraints
• Decentralized Cooperative Stochastic Bandits

Cited by

• Co-Exploration and Co-Exploitation via Shared Structure in Multi-Task Bandits
• Decentralized Contextual Bandits with Network Adaptivity

Related

• Federated Online Sparse Decision Making
• Federated Linear Bandits with Finite Adversarial Actions
• Federated Learning for Heterogeneous Bandits with Unobserved Contexts
• A Simple and Provably Efficient Algorithm for Asynchronous Federated Contextual Linear Bandits
• Federated Linear Contextual Bandits with Heterogeneous Clients