Robust Federated Recommendation System

By Chen Chen, Jingfeng Zhang, Anthony K. H. Tung, Mohan Kankanhalli, Gang Chen. 2020

33 citations recorded by the source corpus.

Federated recommendation systems can provide good performance without collecting users' private data, making them attractive. However, they are susceptible to low-cost poisoning attacks that can degrade their performance. In this paper, we develop a novel federated recommendation technique that is robust against the poisoning attack where Byzantine clients prevail. We argue that the key to Byzantine detection is monitoring of gradients of the model parameters of clients. We then propose a robust learning strategy where instead of using model parameters, the central server computes and utilizes the gradients to filter out Byzantine clients. Theoretically, we justify our robust learning strategy by our proposed definition of Byzantine resilience. Empirically, we confirm the efficacy of our robust learning strategy employing four datasets in a federated recommendation system.

Subfields: Artificial Intelligence, Information Systems

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Qiang Yang
2. Philip S. Yu
3. Lichao Sun
4. Murali Annavaram
5. Chaoyang He
6. Guodong Long
7. Lingjuan Lyu
8. Jing Jiang

In Information Systems

1. Jing Jiang
2. Guodong Long
3. Tianyi Zhou
4. Yue Tan
5. Lu Liu
6. Lingjuan Lyu
7. Chengqi Zhang
8. Philip S. Yu

Citations in the corpus

• FedML: A Research Library and Benchmark for Federated Machine Learning (2020, background): “…computer vision [77, 78], natural language processing [79, 80, 81, 82, 83], finance [74, 35, 84], transportation [85, 86, 87, 88, 89, 90, 91, 92, 93, 94], digital health [95, 96, 97, 98, 99, 100, 101], recommendation [102, 103, 104, 105, 106, 107], robotics [108, 109], and smart cities [110, 111].”
• Privacy and Robustness in Federated Learning: Attacks and Defenses (2020, background): “For example, with Paillier encryption scheme, the encryption of an encoded floating-point number (whether single or double precision) is 2 m bits long, where m is typically at least 1024 and the addition of two encrypted numbers is 2 ∼ 3 orders of magnitude slower than the unencrypted equivalent [9].”
• FedProto: Federated Prototype Learning across Heterogeneous Clients (2021, background): “All of them use gradient-based aggregation methods which will raise concerns about communication efficiency and gradient-based attacks (Zhu, Liu, and Han 2019; Chen et al. 2020; Liu et al. 2021b; Zheng et al. 2021).”