prabhant/synthesizing-robust-adversarial-examples

My entry for ICLR 2018 Reproducibility Challenge for paper Synthesizing robust adversarial examples https://openreview.net/pdf?id=BJDH5M-AW

35
/ 100
Emerging

This project helps machine learning engineers and researchers understand how to create "adversarial examples" that can fool image recognition systems. It takes an image and a target classification, then outputs a modified image that looks normal to humans but is consistently misclassified by AI models, even under various transformations like rotation or cropping. This is valuable for anyone working on the security and reliability of computer vision systems.

No commits in the last 6 months.

Use this if you need to test the robustness of your image recognition models against sophisticated attack methods or explore the vulnerabilities of deep learning systems to crafted inputs.

Not ideal if you are looking for a general-purpose tool to improve the accuracy or performance of your standard image classification tasks.

AI Security Computer Vision Machine Learning Research Adversarial Attacks Model Robustness
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 9 / 25
Maturity 8 / 25
Community 18 / 25

How are scores calculated?

Stars

72

Forks

14

Language

Jupyter Notebook

License

Last pushed

Apr 05, 2018

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/prabhant/synthesizing-robust-adversarial-examples"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.