Summary of Privacy-enhancing Collaborative Information Sharing Through Federated Learning — a Case Of the Insurance Industry, by Panyi Dong et al.
Privacy-Enhancing Collaborative Information Sharing through Federated Learning – A Case of the Insurance Industry
by Panyi Dong, Zhiyu Quan, Brandon Edwards, Shih-han Wang, Runhuan Feng, Tianyang Wang, Patrick Foley, Prashant Shah
First submitted to arxiv on: 22 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Cryptography and Security (cs.CR); Risk Management (q-fin.RM)
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 Federated Learning (FL) approach enables insurance companies to collaborate on building a single model for claims loss modeling without sharing their private data. This addresses concerns around limited data volume and variety caused by privacy issues and rarity of claim events. During each FL round, local models are updated using private data, and the insights are combined to update a global model, leading to improved forecasting accuracy compared to individual model training. The open-source OpenFL framework allows for confidential computing and algorithmic protections against information leakage. This FL-based collaborative learning technique addresses data privacy concerns in traditional machine learning solutions, and its application can be expanded to fraud detection and catastrophe modeling. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Insurance companies can now work together on a single claims loss model without sharing their private data. This is achieved through Federated Learning (FL), which updates local models using private data and combines the insights to create a global model. This approach improves forecasting accuracy compared to individual model training. A special framework called OpenFL allows for confidential computing and protects against information leakage. This technique can be used in other areas, such as fraud detection and catastrophe modeling, where data privacy is important. |
Keywords
* Artificial intelligence * Federated learning * Machine learning
