Summary of Lffr: Logistic Function For (single-output) Regression, by John Chiang
LFFR: Logistic Function For (single-output) Regression
by John Chiang
First submitted to arxiv on: 13 Jul 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 A machine learning paper introduces privacy-preserving regression techniques using fully homomorphic encryption. It proposes simplified Hessians for linear and ridge regressions, enabling efficient training on encrypted data. A novel algorithm called LFFR is developed for homomorphic regression with the logistic function, allowing modeling of complex relationships between input values and output predictions. The algorithm is evaluated on two real-world datasets, and the paper suggests normalizing both data and target predictions to maintain weights in a small range and avoid tuning regularization parameters. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps protect people’s privacy by using special encryption that allows powerful machine learning techniques without revealing individual information. It starts with linear regression, making it work for any dataset, not just ones that are already scaled between 0 and 1. The paper also develops a new way to do ridge regression, which adds a penalty term to keep the model’s weights in check. The main idea is a new algorithm called LFFR that uses the logistic function to predict outcomes based on input values. It tests this algorithm on real-world datasets and suggests normalizing both data and predictions to make it work better. |
Keywords
» Artificial intelligence » Linear regression » Machine learning » Regression » Regularization
