Foolbox

Tool (not an eval — tooling/reference entry)

Adversarial attacks library.

Category: Adversarial ML (classic)
Insurance exposure: Model failure
Publisher: Tübingen
Publisher type: Academic
License: MIT
Status: Stale (>12mo)
Last commit: 2025-12-03
Stars: 2984
Notes: Stale.

View source on GitHub →

Wanich Risk Lab, 2026.