Summary of Jigsaw Game: Federated Clustering, by Jinxuan Xu et al.
Jigsaw Game: Federated Clustering
by Jinxuan Xu, Hong-You Chen, Wei-Lun Chao, Yuqian Zhang
First submitted to arxiv on: 17 Jul 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
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 Federated learning has been gaining attention, particularly within supervised learning domains. However, unsupervised learning problems like clustering in the federated setting remain underexplored. This paper investigates the federated clustering problem, focusing on federated k-means. The challenge lies in the non-convex objective and data heterogeneity in the federated framework. To address these issues, the authors propose a one-shot algorithm called FeCA (Federated Centroid Aggregation). FeCA adaptively refines local solutions on clients, then aggregates these refined solutions to recover the global solution of the entire dataset in a single round. The paper empirically demonstrates the robustness of FeCA under various federated scenarios using synthetic and real-world data. Additionally, the authors extend FeCA to representation learning, introducing DeepFeCA, which combines DeepCluster and FeCA for unsupervised feature learning in the federated setting. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about a new way to do clustering (grouping similar things together) when you have lots of data from different sources. Clustering is important because it helps us understand what our data looks like and how it’s related. The problem is that when we’re doing clustering with lots of different data, it gets really hard. So, the authors came up with a new way to do clustering called FeCA (Federated Centroid Aggregation). It works by taking all the small pieces of data from each source, making them better, and then putting them together in one big piece. The paper shows that this method works well on lots of different kinds of data. |
Keywords
» Artificial intelligence » Attention » Clustering » Federated learning » K means » One shot » Representation learning » Supervised » Unsupervised
