Summary of Adversarial Perturbations Of Physical Signals, by Robert L. Bassett et al.
Adversarial Perturbations of Physical Signals
by Robert L. Bassett, Austin Van Dellen, Anthony P. Austin
First submitted to arxiv on: 27 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Cryptography and Security (cs.CR); Signal Processing (eess.SP); Optimization and Control (math.OC); Machine Learning (stat.ML)
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 investigates the vulnerability of computer-vision-based signal classifiers to adversarial perturbations of their inputs, considering signals and perturbations subject to physical constraints. Specifically, it explores how to construct interfering signals that cause misclassification despite nearly imperceptible perturbations on the received signal’s spectrogram, using pre-trained neural networks. The authors introduce methods for solving PDE-constrained optimization problems efficiently, leading to effective and physically realizable adversarial perturbations for various machine learning models under different physical conditions. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper looks at how machine learning models can be tricked into making wrong decisions by adding tiny changes to the signals they receive. It’s like trying to fool a person into thinking one thing is another thing, even if it’s just slightly different. The researchers used special mathematical problems to create these tricks and tested them on different kinds of computer vision models. They found ways to make these tricks work efficiently, which could be useful in fields like security or self-driving cars. |
Keywords
* Artificial intelligence * Machine learning * Optimization
