Summary of Agent-oriented Joint Decision Support For Data Owners in Auction-based Federated Learning, by Xiaoli Tang et al.
Agent-oriented Joint Decision Support for Data Owners in Auction-based Federated Learning
by Xiaoli Tang, Han Yu, Xiaoxiao Li
First submitted to arxiv on: 9 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Computer Science and Game Theory (cs.GT)
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 approach to Auction-based Federated Learning (AFL), which motivates data owners to participate in federated learning through economic means. The proposed agent-oriented joint Pricing, Acceptance and Sub-delegation decision support approach for data owners in AFL (PAS-AFL) provides a systematic way for data owners to make decisions on bid acceptance, task sub-delegation, and pricing to maximize their utility using Lyapunov optimization. This approach allows each data owner to take on multiple federated learning tasks simultaneously, enhancing the throughput of FL tasks in the AFL ecosystem. The authors demonstrate the effectiveness of PAS-AFL by comparing it to six alternative strategies on six benchmarking datasets, showing significant advantages in terms of utility and test accuracy. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Auction-based Federated Learning is a way for many devices to work together and learn from each other’s data without sharing their personal information. The problem is that some devices don’t have enough motivation to participate. This paper proposes a new way to make decisions about which tasks to take on, how much money to charge for participating, and whether to give up control of certain tasks. It uses math called Lyapunov optimization to help devices make the best choices for themselves. The results show that this approach can increase the overall performance of the learning process by 28.77% and 2.64%. |
Keywords
» Artificial intelligence » Federated learning » Optimization
