Summary of Federated Progressive Self-distillation with Logits Calibration For Personalized Iiot Edge Intelligence, by Yingchao Wang and Wenqi Niu
Federated Progressive Self-Distillation with Logits Calibration for Personalized IIoT Edge Intelligence
by Yingchao Wang, Wenqi Niu
First submitted to arxiv on: 30 Nov 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- 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 This study proposes a novel Personalized Federated Learning (PFL) method called Federated Progressive Self-Distillation (FedPSD), which addresses the issue of forgetting both historical personalized knowledge and global generalized knowledge during local training on clients. The proposed approach combines logits calibration, progressive self-distillation, and calibrated fusion labels to tackle data heterogeneity and diverse user needs in industrial IoT (IIoT) settings. FedPSD leverages virtual teachers from previous epochs to guide the training of subsequent epochs, enabling rapid recall of personalized knowledge. Experimental results demonstrate the effectiveness and superiority of the proposed method under various data heterogeneity scenarios. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about making sure that machines learning models remember important information even when they’re trained separately on different devices. Right now, these models can forget what they learned before, which means they might not work as well when they need to recall previous knowledge. The researchers came up with a new way to train these models called Federated Progressive Self-Distillation (FedPSD). It uses a combination of techniques to help the models remember both personal and general information. They tested this method on different types of data and found that it worked really well. |
Keywords
» Artificial intelligence » Distillation » Federated learning » Logits » Recall
