Summary of Dataset Distillation-based Hybrid Federated Learning on Non-iid Data, by Xiufang Shi et al.
Dataset Distillation-based Hybrid Federated Learning on Non-IID Data
by Xiufang Shi, Wei Zhang, Mincheng Wu, Guangyi Liu, Zhenyu Wen, Shibo He, Tejal Shah, Rajiv Ranjan
First submitted to arxiv on: 26 Sep 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 The paper proposes a hybrid federated learning framework called HFLDD to address label distribution skew in non-independently and identically distributed (Non-IID) data. The framework integrates dataset distillation to generate approximately independent and equally distributed (IID) data, improving model training performance. The method partitions clients into heterogeneous clusters with unbalanced labels within each cluster, while balancing labels across clusters. Cluster headers collect distilled data from members and train models collaboratively with the server, alleviating the impact of Non-IID data on model training. Experimental results show that HFLDD outperforms baseline methods in terms of test accuracy and communication cost when data labels are severely imbalanced. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper solves a problem in machine learning where different devices have different types of information. This makes it hard for the machines to work together to learn new things. The researchers created a new way to make sure all the devices can share their information, even if they’re different. They call this method HFLDD, which stands for Hybrid Federated Learning with Dataset Distillation. The method works by putting similar devices together and having them share their information with each other. This helps the machines learn better and work more efficiently. |
Keywords
» Artificial intelligence » Distillation » Federated learning » Machine learning
