Summary of Dpvs-shapley:faster and Universal Contribution Evaluation Component in Federated Learning, by Ketin Yin et al.
DPVS-Shapley:Faster and Universal Contribution Evaluation Component in Federated Learning
by Ketin Yin, Zonghao Guo, ZhengHan Qin
First submitted to arxiv on: 19 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); 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 In a decentralized learning paradigm called federated learning, researchers have been exploring ways to ensure that each participant’s contributions are fairly evaluated and accurately assessed. This approach not only addresses data privacy concerns but also enhances the system’s scalability and robustness. To achieve this, developing an effective contribution evaluation mechanism is crucial. The proposed Dynamic Pruning Validation Set Shapley (DPVS-Shapley) method accelerates the process by dynamically pruning the original dataset without compromising accuracy, allowing clients to receive higher scores based on their ability to distinguish difficult examples. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Federated learning is a new way of training artificial intelligence models that doesn’t require sharing all your data. This is important for privacy reasons. To make sure each part takes an equal role in the process, we need a good way to measure how much each one contributes. A method called Shapley value-based is often used, but it can be slow and not very practical. Our new method, DPVS-Shapley, makes this process faster and more accurate by reducing the amount of data needed. |
Keywords
» Artificial intelligence » Federated learning » Pruning
