Summary of Starlit: Privacy-preserving Federated Learning to Enhance Financial Fraud Detection, by Aydin Abadi et al.
Starlit: Privacy-Preserving Federated Learning to Enhance Financial Fraud Detection
by Aydin Abadi, Bradley Doyle, Francesco Gini, Kieron Guinamard, Sasi Kumar Murakonda, Jack Liddell, Paul Mellor, Steven J. Murdoch, Mohammad Naseri, Hector Page, George Theodorakopoulos, Suzanne Weller
First submitted to arxiv on: 19 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 This paper introduces Starlit, a novel scalable privacy-preserving Federated Learning (FL) mechanism that addresses limitations in existing FL solutions for identifying fraudulent financial transactions. These limitations include the lack of formal security definitions and proofs, assumption of prior account freezing, poor scaling, and exclusion of identity alignment phase from implementation and analysis. Starlit overcomes these issues by leveraging modular arithmetic to achieve efficient computation while maintaining privacy. The paper presents a thorough performance evaluation using synthetic data from a key global financial transactions player, demonstrating Starlit’s scalability, efficiency, and accuracy. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Starlit is a new way for banks and other financial institutions to work together to catch scammers without sharing their customers’ personal information. Currently, there are problems with how this works now, like not having a clear definition of what “secure” means or assuming that the bad guys have already been caught before we start working on it. This makes it hard for banks to use these methods and keeps them from being as good as they could be. Starlit fixes all these problems by using special math tricks to make sure everything is private and secure. It’s like a superpower that helps catch scammers better and faster. |
Keywords
* Artificial intelligence * Alignment * Federated learning * Synthetic data
