Summary of Enhancing Adversarial Transferability Via Information Bottleneck Constraints, by Biqing Qi et al.
Enhancing Adversarial Transferability via Information Bottleneck Constraints
by Biqing Qi, Junqi Gao, Jianxing Liu, Ligang Wu, Bowen Zhou
First submitted to arxiv on: 8 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- 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 paper proposes a novel framework, IBTA, for generating transferable adversarial attacks that leverages invariant features. This approach aims to reduce the reliance of perturbations on original data and instead focus on invariant features critical to classification. The authors redefine the optimization process using an information bottleneck (IB) theory-based framework, which addresses the challenge of unoptimizable mutual information. They introduce a simple and efficient mutual information lower bound (MILB) and utilize the Mutual Information Neural Estimator (MINE) for analysis. Experiments on the ImageNet dataset demonstrate the efficiency and scalability of IBTA and MILB. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper explores ways to make attacks work better across different situations. It suggests a new approach called IBTA that uses important features that don’t change much, even when the data is modified. This helps the attack be more effective in other scenarios. The authors also came up with a way to estimate how well this approach works and tested it on many images. |
Keywords
* Artificial intelligence * Classification * Optimization
