Q-TART: Quickly Training for Adversarial Robustness and in-Transferability

TL;DR

Q-TART is a training method that enhances DNN performance, efficiency, and adversarial robustness by removing noise-susceptible samples, achieving better results with less training data and time.

Contribution

The paper introduces Q-TART, a novel algorithm that improves robustness and efficiency by selectively removing noise-sensitive samples during training.

Findings

Improves adversarial robustness by over 1.3%.

Reduces training time by up to 17.9%.

Effective across multiple datasets including ImageNet.

Abstract

Raw deep neural network (DNN) performance is not enough; in real-world settings, computational load, training efficiency and adversarial security are just as or even more important. We propose to simultaneously tackle Performance, Efficiency, and Robustness, using our proposed algorithm Q-TART, Quickly Train for Adversarial Robustness and in-Transferability. Q-TART follows the intuition that samples highly susceptible to noise strongly affect the decision boundaries learned by DNNs, which in turn degrades their performance and adversarial susceptibility. By identifying and removing such samples, we demonstrate improved performance and adversarial robustness while using only a subset of the training data. Through our experiments we highlight Q-TART's high performance across multiple Dataset-DNN combinations, including ImageNet, and provide insights into the complementary behavior of Q-TART alongside existing adversarial training approaches to increase robustness by over 1.3% while using up to 17.9% less training time.