Summary of Fedsc: Provable Federated Self-supervised Learning with Spectral Contrastive Objective Over Non-i.i.d. Data, by Shusen Jing et al.
FedSC: Provable Federated Self-supervised Learning with Spectral Contrastive Objective over Non-i.i.d. Data
by Shusen Jing, Anlan Yu, Shuai Zhang, Songyang Zhang
First submitted to arxiv on: 7 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Cryptography and Security (cs.CR); Signal Processing (eess.SP)
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 The paper proposes a new federated self-supervised learning (FedSSL) algorithm, named FedSC, which integrates self-supervised learning with federated learning. FedSC uses spectral contrastive objectives to minimize the global objective of FedSSL and improve data representation quality. The algorithm also incorporates differential privacy protection to control additional privacy leakage when sharing correlation matrices between clients. The authors provide theoretical analysis on convergence and extra privacy leakage. Experimental results validate the effectiveness of FedSC. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary FedSSL combines self-supervised learning with federated learning, but conventional approaches like federated averaging fail to minimize the global objective. The new algorithm, FedSC, uses spectral contrastive objectives to improve data representation quality. It also includes differential privacy protection to control additional privacy leakage when sharing correlation matrices between clients. This helps ensure better performance and protects user data. |
Keywords
» Artificial intelligence » Federated learning » Self supervised
