Summary of Share Your Secrets For Privacy! Confidential Forecasting with Vertical Federated Learning, by Aditya Shankar et al.
Share Your Secrets for Privacy! Confidential Forecasting with Vertical Federated Learning
by Aditya Shankar, Lydia Y. Chen, Jérémie Decouchant, Dimitra Gkorou, Rihan Hai
First submitted to arxiv on: 31 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Cryptography and Security (cs.CR); Distributed, Parallel, and Cluster Computing (cs.DC)
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 ‘Secret-shared Time Series Forecasting with VFL’ (STV) framework addresses the challenges of data privacy, over-fitting on small and noisy datasets, and scalability in vertical federated learning for industrial applications. STV combines SARIMAX and autoregressive trees on vertically partitioned data, secret sharing, and multi-party computation for serverless forecasting. Novel N-party algorithms are used for matrix multiplication and inverse operations to optimize model parameters directly. Evaluations on six representative datasets demonstrate comparable accuracy to centralized approaches and outperformance by 23.81% in some cases. A scalability analysis examines communication costs of direct and iterative optimization. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary In this paper, researchers created a new way to do time series forecasting using vertical federated learning. This is useful for industries that want to predict what will happen next based on past data. They had to overcome some big challenges like keeping data private and making sure the model works well even with small amounts of noisy data. The new method, called STV, uses special algorithms and math to make predictions without sharing all the data. The results show that STV is just as good as more traditional methods in many cases. The researchers also looked at how well their method could handle big datasets and found that it was faster and more efficient than some other approaches. |
Keywords
» Artificial intelligence » Autoregressive » Federated learning » Optimization » Time series
