Summary of Vidmodex: Interpretable and Efficient Black Box Model Extraction For High-dimensional Spaces, by Somnath Sendhil Kumar et al.
VidModEx: Interpretable and Efficient Black Box Model Extraction for High-Dimensional Spaces
by Somnath Sendhil Kumar, Yuvaraj Govindarajulu, Pavan Kulkarni, Manojkumar Parmar
First submitted to arxiv on: 4 Aug 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
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 presents a novel approach to black-box model extraction using SHAP (SHapley Additive exPlanations) to enhance synthetic data generation. The method leverages the individual contributions of each input feature towards the victim model’s output, as quantified by SHAP, to optimize an energy-based GAN towards a desirable output. This innovative technique significantly boosts performance, achieving a 16.45% increase in image classification models and extending to video classification models with average improvements of 26.11% and maximum gains of 33.36% on challenging datasets such as UCF11, UCF101, Kinetics 400, Kinetics 600, and Something-Something V2. The method’s effectiveness is demonstrated under various scenarios, including the availability of top-k prediction probabilities, top-k prediction labels, and top-1 labels. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper introduces a new way to analyze how a “black box” computer program makes decisions. The researchers use a technique called SHAP to understand how each part of the input contributes to the program’s output. This helps generate fake data that is similar to real data, which improves the accuracy of image and video classification models by 16-33%. The method works well even when only partial information about the program’s decisions is available. |
Keywords
* Artificial intelligence * Classification * Gan * Image classification * Synthetic data
