Summary of Adafed: Fair Federated Learning Via Adaptive Common Descent Direction, by Shayan Mohajer Hamidi et al.
AdaFed: Fair Federated Learning via Adaptive Common Descent Direction
by Shayan Mohajer Hamidi, En-Hui Yang
First submitted to arxiv on: 10 Jan 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
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 tackle a critical problem in Federated Learning (FL), where trained models may unfairly advantage or disadvantage some devices. They propose AdaFed, a method to find an updating direction for the server that ensures all clients’ loss functions decrease and those with larger values decrease at a higher rate. This is achieved by adaptively tuning the common direction based on local gradients and loss functions. The authors demonstrate the effectiveness of AdaFed on a range of federated datasets, outperforming state-of-the-art fair FL methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Federated learning helps devices work together to train machine learning models. But sometimes, this training can be unfair, making some devices better or worse off than others. To fix this, scientists created a new method called AdaFed. It’s like finding the best way for all devices to move in the same direction, so everyone gets an equal chance. They tested AdaFed and found it worked better than other methods. |
Keywords
* Artificial intelligence * Federated learning * Machine learning
