Federated Linear Contextual Bandits with User-level Differential Privacy
Authors: Ruiquan Huang, Huanyu Zhang, Luca Melis, Milan Shen, Meisam Hejazinia, Jing Yang
ICML 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | This paper studies federated linear contextual bandits under the notion of user-level differential privacy (DP). We first introduce a unified federated bandits framework that can accommodate various definitions of DP in the sequential decisionmaking setting. We then formally introduce userlevel central DP (CDP) and local DP (LDP) in the federated bandits framework, and investigate the fundamental trade-offs between the learning regrets and the corresponding DP guarantees in a federated linear contextual bandits model. For CDP, we propose a federated algorithm termed as ROBIN and show that it is near-optimal in terms of the number of clients M and the privacy budget ε by deriving nearly-matching upper and lower regret bounds when user-level DP is satisfied. For LDP, we obtain several lower bounds, indicating that learning under user-level (ε, δ)-LDP must suffer a regret blow-up factor at least min{1/ε, M} or min{1/ ε, M} under different conditions. |
| Researcher Affiliation | Collaboration | 1School of EECS, The Pennsylvania State University, University Park, PA, USA 2Meta, USA 3Google, USA. |
| Pseudocode | Yes | Algorithm 1 The ROBIN Algorithm |
| Open Source Code | No | The paper does not provide any explicit statement about releasing source code or a link to a code repository for the methodology described. |
| Open Datasets | No | The paper is theoretical and does not involve training models on datasets. It describes a theoretical framework for federated linear contextual bandits. |
| Dataset Splits | No | The paper is theoretical and does not conduct experiments with dataset splits. It provides theoretical bounds and analyses. |
| Hardware Specification | No | The paper is theoretical and does not conduct experiments, thus no hardware specifications are mentioned. |
| Software Dependencies | No | The paper is theoretical and does not conduct experiments, thus no software dependencies with version numbers are listed. |
| Experiment Setup | No | The paper is theoretical and does not describe an experimental setup with hyperparameters or training configurations. |