Most existing machine learning classifiers are highly vulnerable to adversarial examples. An adversarial example is a sample of input data which has been modified very slightly in a way that is intended to cause a machine learning classifier to misclassify it.

To accelerate research on adversarial examples, GeekPwn was partnering with Alexey Kurakin from Google Brain and Dawn Song from UC Berkeley EECS to organize Competition on Adversarial Attacks and Defenses 2018 (CAAD2018). Submittions started at May 10th, 2018 and ended at Aug. 31th, 2018.

There are 3 sub-competitions: Targeted attack, Non-targeted attack and Defense. 1st place for each sub-competition wins RMB100K, 2nd place wins RMB55K, 3rd place wins RMB35K, 4th and 5th places win RMB6K。