Summary of One-shot Sequential Federated Learning For Non-iid Data by Enhancing Local Model Diversity, By Naibo Wang et al.
One-Shot Sequential Federated Learning for Non-IID Data by Enhancing Local Model Diversity
by Naibo Wang, Yuchen Deng, Wenjie Feng, Shichen Fan, Jianwei Yin, See-Kiong Ng
First submitted to arxiv on: 18 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Computer Vision and Pattern Recognition (cs.CV); 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 Medium Difficulty summary: This paper proposes a novel approach to improve one-shot sequential federated learning for non-independent and identically distributed data. The authors introduce a local model pool that stores diverse models generated during local training, and two distance measurements to enhance model diversity and mitigate the effects of non-IID data. Their proposed framework achieves better accuracy compared to state-of-the-art methods on both label-skew and domain-shift tasks, such as 6%+ accuracy improvement on the CIFAR-10 dataset. The authors demonstrate their method’s superiority over existing one-shot parallel federated learning methods. This research has implications for improving model performance while maintaining low communication costs in distributed machine learning settings. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Low Difficulty summary: This paper is about making it easier to learn from lots of different computers without sharing too much data. Right now, when we do this, it can be slow and not very accurate. The authors came up with a new way to make it faster and better by storing many different models on each computer and comparing them to get the best one. They tested their idea and found that it works really well, especially for difficult problems like recognizing pictures of animals or objects. This could help us learn from lots of computers without sharing too much data. |
Keywords
» Artificial intelligence » Federated learning » Machine learning » One shot
