Get Fooled for the Right Reason: Improving Adversarial Robustness through a Teacher-guided Curriculum Learning Approach

10/30/2021
by   Anindya Sarkar, et al.
0

Current SOTA adversarially robust models are mostly based on adversarial training (AT) and differ only by some regularizers either at inner maximization or outer minimization steps. Being repetitive in nature during the inner maximization step, they take a huge time to train. We propose a non-iterative method that enforces the following ideas during training. Attribution maps are more aligned to the actual object in the image for adversarially robust models compared to naturally trained models. Also, the allowed set of pixels to perturb an image (that changes model decision) should be restricted to the object pixels only, which reduces the attack strength by limiting the attack space. Our method achieves significant performance gains with a little extra effort (10-20 terms of adversarial as well as natural accuracy. We have performed extensive experimentation with CIFAR-10, CIFAR-100, and TinyImageNet datasets and reported results against many popular strong adversarial attacks to prove the effectiveness of our method.

READ FULL TEXT

page 8

page 9

page 15

page 16

research
11/03/2018

Learning to Defense by Learning to Attack

Adversarial training provides a principled approach for training robust ...
research
02/14/2020

Adversarial Distributional Training for Robust Deep Learning

Adversarial training (AT) is among the most effective techniques to impr...
research
05/29/2019

Functional Adversarial Attacks

We propose functional adversarial attacks, a novel class of threat model...
research
04/25/2020

Improved Adversarial Training via Learned Optimizer

Adversarial attack has recently become a tremendous threat to deep learn...
research
05/31/2021

Adversarial Training with Rectified Rejection

Adversarial training (AT) is one of the most effective strategies for pr...
research
06/13/2021

ATRAS: Adversarially Trained Robust Architecture Search

In this paper, we explore the effect of architecture completeness on adv...

Please sign up or login with your details

Forgot password? Click here to reset