Summary of Is My Data in Your Ai Model? Membership Inference Test with Application to Face Images, by Daniel Dealcala et al.
Is my Data in your AI Model? Membership Inference Test with Application to Face Images
by Daniel DeAlcala, Aythami Morales, Julian Fierrez, Gonzalo Mancera, Ruben Tolosana, Javier Ortega-Garcia
First submitted to arxiv on: 14 Feb 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
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 Membership Inference Test (MINT) is a novel approach designed to assess whether given data was used during the training of AI/ML models. Two MINT architectures, based on Multilayer Perceptrons (MLPs) and Convolutional Neural Networks (CNNs), learn distinct activation patterns when an Audited Model is exposed to data used during its training process. The experimental framework focuses on Face Recognition, considering three state-of-the-art systems, and uses six publicly available databases with over 22 million face images. The proposed MINT approach achieves promising results, with up to 90% accuracy, indicating the potential to recognize if an AI model has been trained with specific data. This can serve to enforce privacy and fairness in several AI applications, such as revealing if sensitive or private data was used for training or tuning Large Language Models (LLMs). |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The Membership Inference Test is a new way to check if certain data was used to train artificial intelligence models. It uses special computer programs called Multilayer Perceptrons and Convolutional Neural Networks to look at how the model acts when it sees this training data again. The test focuses on recognizing faces, using three good systems and many face pictures. This new way of testing works well, getting up to 90% correct! This can help keep private information safe in things like language learning models. |
Keywords
» Artificial intelligence » Face recognition » Inference
