Summary of Approximate Gradient Coding For Privacy-flexible Federated Learning with Non-iid Data, by Okko Makkonen et al.
Approximate Gradient Coding for Privacy-Flexible Federated Learning with Non-IID Data
by Okko Makkonen, Sampo Niemelä, Camilla Hollanti, Serge Kas Hanna
First submitted to arxiv on: 4 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Cryptography and Security (cs.CR); Distributed, Parallel, and Cluster Computing (cs.DC); Information Theory (cs.IT); Machine Learning (stat.ML)
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 tackles the challenges of non-independent and identically distributed (non-IID) data and stragglers/dropouts in federated learning. It proposes a privacy-flexible paradigm that models parts of clients’ local data as non-private, offering a more versatile and business-oriented perspective on privacy. The authors introduce a data-driven strategy for mitigating the effects of label heterogeneity and client straggling on federated learning, combining offline data sharing and approximate gradient coding techniques. Numerical simulations using the MNIST dataset demonstrate that this approach enables achieving a deliberate trade-off between privacy and utility, leading to improved model convergence and accuracy while using an adaptable portion of non-private data. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps solve problems in a special kind of learning called federated learning. Federated learning is when many devices work together to learn something new without sharing all their data with each other. The problem is that the devices might have different types of data, and some devices might not be working properly. To fix this, the authors came up with a way to make it more flexible and let devices share parts of their data if they want to. They also created a new way to deal with differences in the data labels and when some devices are slower than others. The results show that this approach makes it possible to balance privacy and usefulness while still getting good results. |
Keywords
* Artificial intelligence * Federated learning
