Shared References

Micah Goldblum, Dimitris Tsipras, Chulin Xie, Xinyun Chen, Avi Schwarzschild, Dawn Song, Aleksander Madry, Bo Li, Tom Goldstein

Comprehensive survey systematically categorizing dataset vulnerabilities including poisoning and backdoor attacks, their threat models, and defense mechanisms.

Battista Biggio, Blaine Nelson, Pavel Laskov

Investigates poisoning attacks against SVMs where adversaries inject crafted training data to increase test error. Uses gradient ascent to construct malicious data points.