Summary of Personalized Federated Learning Of Probabilistic Models: a Pac-bayesian Approach, by Mahrokh Ghoddousi Boroujeni et al.
Personalized Federated Learning of Probabilistic Models: A PAC-Bayesian Approach
by Mahrokh Ghoddousi Boroujeni, Andreas Krause, Giancarlo Ferrari Trecate
First submitted to arxiv on: 16 Jan 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Cryptography and Security (cs.CR)
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 Personalized Federated Learning algorithm, PAC-PFL, aims to infer a shared probabilistic model from private data stored locally by multiple clients. By utilizing differential privacy and a PAC-Bayesian framework, the algorithm collaboratively learns a shared hyper-posterior and adapts it to each client’s posterior inference for personalization. This approach effectively combats over-fitting by establishing and minimizing a generalization bound on the average true risk of clients. The algorithm achieves accurate and well-calibrated predictions on various datasets, including photovoltaic panel power generation, FEMNIST, and Dirichlet-partitioned EMNIST. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary PAC-PFL is a new way to learn models that work better for different people or things. It does this by using information from many small datasets instead of just one big one. This helps make the model more accurate and personalized. The algorithm also makes sure the predictions are reliable and not too confident. Scientists tested it on several datasets, including ones about solar panels and handwritten letters. |
Keywords
* Artificial intelligence * Federated learning * Generalization * Inference * Probabilistic model
