Summary of Fedsat: a Statistical Aggregation Approach For Class Imbalanced Clients in Federated Learning, by Sujit Chowdhury et al.
FedSat: A Statistical Aggregation Approach for Class Imbalanced Clients in Federated Learning
by Sujit Chowdhury, Raju Halder
First submitted to arxiv on: 4 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 This paper proposes a novel federated learning (FL) approach called FedSat, which addresses three forms of data heterogeneity: label skewness, missing classes, and quantity skewness. The approach combines a prediction-sensitive loss function with a prioritized-class based weighted aggregation scheme to enhance model performance on minority classes. Experimental results across diverse data-heterogeneity settings demonstrate that FedSat outperforms state-of-the-art baselines by an average of 1.8%, with faster convergence compared to existing methods. This highlights the effectiveness of FedSat in addressing heterogeneous FL challenges and its potential for real-world applications. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary FedSat is a new way for devices to learn together while keeping their data private. The problem is that devices have different kinds of information, which makes it hard to work together effectively. The solution involves two main parts: a special way to measure how good the model is and a way to combine the devices’ contributions based on what they’re good at. This approach worked better than other methods in tests and can be used in real-life applications. |
Keywords
» Artificial intelligence » Federated learning » Loss function
