Summary of Provable Privacy Attacks on Trained Shallow Neural Networks, by Guy Smorodinsky and Gal Vardi and Itay Safran
Provable Privacy Attacks on Trained Shallow Neural Networks
by Guy Smorodinsky, Gal Vardi, Itay Safran
First submitted to arxiv on: 10 Oct 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 investigates the privacy threats against 2-layer ReLU neural networks, focusing on data reconstruction and membership inference attacks. The authors demonstrate how theoretical results on implicit bias can be leveraged to reconstruct a subset of training points with high probability in low-dimensional settings, and identify whether a given point was used in training with high accuracy in high-dimensional settings. This study showcases the first provable vulnerabilities in this context, highlighting the need for robust privacy measures. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research looks at how well-trained neural networks can be attacked to reveal sensitive information. The team finds ways to reconstruct what’s inside a network and figure out if new data comes from their original training set or not. This is important because it shows that these networks are not as secure as we thought, and we need better protection against privacy threats. |
Keywords
» Artificial intelligence » Inference » Probability » Relu
