Summary of Towards Fair, Robust and Efficient Client Contribution Evaluation in Federated Learning, by Meiying Zhang et al.
Towards Fair, Robust and Efficient Client Contribution Evaluation in Federated Learning
by Meiying Zhang, Huan Zhao, Sheldon Ebron, Kan Yang
First submitted to arxiv on: 6 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); Distributed, Parallel, and Cluster Computing (cs.DC)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes Fair, Robust, and Efficient Client Assessment (FRECA), a novel method for evaluating client contributions in Federated Learning (FL) systems. FRECA tackles the challenge of non-independent and identically distributed (non-iid) data by introducing FedTruth, a framework that estimates the global model’s ground truth update while filtering out malicious updates. The approach is robust against Byzantine attacks and efficient, as it only requires local model updates and no validation operations or datasets. Experimental results demonstrate FRECA’s ability to accurately and efficiently quantify client contributions in FL systems. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary In Federated Learning (FL), clients’ performance can vary due to different reasons. It’s important to know how each client contributes to the overall learning process, so that you can choose which clients are best for your project. This paper introduces a new way to measure client contributions called FRECA. FRECA helps by estimating what the global model should look like and filtering out any bad information from malicious clients. This approach is good because it’s not affected by fake data and doesn’t require extra datasets or validation steps. |
Keywords
* Artificial intelligence * Federated learning
