How Adversarial Attacks Work. Adversarial attacks against machine learning classifiers involve constructing an input that deliberately produces the wrong classification. This article shows how these can be constructed, and includes examples generated using PyTorch which produce a sports car that gets identified as a toaster and a photo of Sylvester Stallone that gets classified as Keanu Reeves.
Recent articles
- Claude Opus 4.5, and why evaluating new LLMs is increasingly difficult - 24th November 2025
- sqlite-utils 4.0a1 has several (minor) backwards incompatible changes - 24th November 2025
- Olmo 3 is a fully open LLM - 22nd November 2025