Summary of Taming Cross-domain Representation Variance in Federated Prototype Learning with Heterogeneous Data Domains, by Lei Wang et al.
Taming Cross-Domain Representation Variance in Federated Prototype Learning with Heterogeneous Data Domains
by Lei Wang, Jieming Bian, Letian Zhang, Chen Chen, Jie Xu
First submitted to arxiv on: 14 Mar 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Computer Vision and Pattern Recognition (cs.CV)
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 proposed Federated Prototype Learning with Variance-aware dual-level prototypes clustering (FedPLVM) improves federated learning by addressing the issue of heterogeneous data domains across clients. The approach creates local clustered prototypes based on private data features and then performs global prototypes clustering to reduce communication complexity and preserve local data privacy. A novel α-sparsity prototype loss is introduced to align samples from underrepresented domains, enhancing intra-class similarity and reducing inter-class similarity. This method outperforms existing approaches on Digit-5, Office-10, and DomainNet datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Federated learning helps computers learn together without sharing personal data. But in real-life situations, different devices often have different types of data. A new approach called Federated Prototype Learning (FedPL) tries to solve this problem by using special examples, or “prototypes,” to help models work better with diverse data. However, existing methods for FedPL don’t always create the right number of prototypes for each device, leading to differences in how well they perform. The new method, FedPLVM, uses a unique way of grouping prototypes to reduce the difference and preserve privacy. |
Keywords
* Artificial intelligence * Clustering * Federated learning
