Summary of Pursuing Overall Welfare in Federated Learning Through Sequential Decision Making, by Seok-ju Hahn et al.
Pursuing Overall Welfare in Federated Learning through Sequential Decision Making
by Seok-Ju Hahn, Gi-Soo Kim, Junghye Lee
First submitted to arxiv on: 31 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Distributed, Parallel, and Cluster Computing (cs.DC); 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 In this paper, researchers tackle the challenge of achieving client-level fairness in federated learning systems. The traditional approach uses a single global model that may not perform equally well for all clients. To address this issue, the authors propose modifying the aggregation scheme to an adaptive one that responds to local signals from participating clients. They unify existing fairness-aware aggregation strategies into an online convex optimization framework, which enables the central server to make sequential decisions. The researchers also introduce AAggFF, a simple and intuitive method for improving suboptimal designs within existing methods. To accommodate practical requirements, they develop two tailored approaches for cross-device and cross-silo settings, respectively. Theoretical analyses show that both settings achieve sublinear regret upper bounds. Experimental results demonstrate that the federated system with AAggFF outperforms existing methods in achieving client-level fairness. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper focuses on making sure a global model is fair to all individual clients when learning from their data together. The current way of doing this might not work well for all clients, so researchers came up with a new approach that adjusts the model based on what each client says. They also found a way to combine different ideas about fairness into one framework. To make it more practical, they created two versions of their method, one for when devices are connected and one for when big groups of devices are working together. The results show that their new approach works better than the old one. |
Keywords
» Artificial intelligence » Federated learning » Optimization
