Reducing Certified Regression to Certified Classification for General Poisoning Attacks

Understand

Adversarial training instances can severely distort a model's behavior.

• This work investigates certified regression defenses, which provide guaranteed limits on how much a regressor's prediction may change under a poisoning attack.
• Our key insight is that certified regression reduces to voting-based certified classification when using median as a model's primary decision function.
• Coupling our reduction with existing certified classifiers, we propose six new regressors provably-robust to poisoning attacks.