Summary of Signsgd with Federated Defense: Harnessing Adversarial Attacks Through Gradient Sign Decoding, by Chanho Park et al.
SignSGD with Federated Defense: Harnessing Adversarial Attacks through Gradient Sign Decoding
by Chanho Park, Namyoon Lee
First submitted to arxiv on: 2 Feb 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 In this paper, researchers explore a novel approach to distributed learning called signSGD with federated defense (signSGD-FD), which enhances the performance of signSGD with majority voting (signSGD-MV) in the presence of adversarial workers. The traditional signSGD-MV algorithm reduces communication costs by one-bit quantization, but its convergence rate decreases as the number of adversarial workers increases. In contrast, signSGD-FD leverages gradient information sent by adversarial workers with proper weights obtained through gradient sign decoding. Experimental results show that signSGD-FD achieves superior convergence rates over traditional algorithms in various adversarial attack scenarios. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary In this paper, scientists develop a new way to make distributed learning work better when some workers might be trying to trick the system. They created an algorithm called signSGD with federated defense (signSGD-FD) that can handle these bad actors. This is important because in real-life situations, not all workers are trustworthy. The researchers tested their algorithm and found it works better than others in different kinds of attacks. |
Keywords
* Artificial intelligence * Quantization
