https://github.com/eth-sri/diffai

A certifiable defense against adversarial examples by training neural networks to be provably robust
https://github.com/eth-sri/diffai

abstract-interpretation attack defense neural-network pytorch robust

Last synced: 4 months ago
JSON representation

A certifiable defense against adversarial examples by training neural networks to be provably robust

Host: GitHub
URL: https://github.com/eth-sri/diffai
Owner: eth-sri
License: mit
Created: 2018-07-08T16:59:42.000Z (over 7 years ago)
Default Branch: master
Last Pushed: 2021-01-13T08:49:14.000Z (almost 5 years ago)
Last Synced: 2024-05-08T00:15:48.712Z (over 1 year ago)
Topics: abstract-interpretation, attack, defense, neural-network, pytorch, robust
Language: Python
Homepage: https://arxiv.org/abs/1903.12519
Size: 1.23 MB
Stars: 218
Watchers: 16
Forks: 26
Open Issues: 1

ecosyste.ms