Adversarial Coreset Selection for Efficient Robust Training

by   Hadi M. Dolatabadi, et al.

Neural networks are vulnerable to adversarial attacks: adding well-crafted, imperceptible perturbations to their input can modify their output. Adversarial training is one of the most effective approaches to training robust models against such attacks. Unfortunately, this method is much slower than vanilla training of neural networks since it needs to construct adversarial examples for the entire training data at every iteration. By leveraging the theory of coreset selection, we show how selecting a small subset of training data provides a principled approach to reducing the time complexity of robust training. To this end, we first provide convergence guarantees for adversarial coreset selection. In particular, we show that the convergence bound is directly related to how well our coresets can approximate the gradient computed over the entire training data. Motivated by our theoretical analysis, we propose using this gradient approximation error as our adversarial coreset selection objective to reduce the training set size effectively. Once built, we run adversarial training over this subset of the training data. Unlike existing methods, our approach can be adapted to a wide variety of training objectives, including TRADES, ℓ_p-PGD, and Perceptual Adversarial Training. We conduct extensive experiments to demonstrate that our approach speeds up adversarial training by 2-3 times while experiencing a slight degradation in the clean and robust accuracy.


page 1

page 2

page 3

page 4


ℓ_∞-Robustness and Beyond: Unleashing Efficient Adversarial Training

Neural networks are vulnerable to adversarial attacks: adding well-craft...

Reducing Adversarial Training Cost with Gradient Approximation

Deep learning models have achieved state-of-the-art performances in vari...

Do we need entire training data for adversarial training?

Deep Neural Networks (DNNs) are being used to solve a wide range of prob...

Friendly Training: Neural Networks Can Adapt Data To Make Learning Easier

In the last decade, motivated by the success of Deep Learning, the scien...

Adversarial Unlearning: Reducing Confidence Along Adversarial Directions

Supervised learning methods trained with maximum likelihood objectives o...

Partitioned Gradient Matching-based Data Subset Selection for Compute-Efficient Robust ASR Training

Training state-of-the-art ASR systems such as RNN-T often has a high ass...

Exploring Memorization in Adversarial Training

It is well known that deep learning models have a propensity for fitting...

Please sign up or login with your details

Forgot password? Click here to reset