Multitask Learning Strengthens Adversarial Robustness

07/14/2020
by   Chengzhi Mao, et al.
31

Although deep networks achieve strong accuracy on a range of computer vision benchmarks, they remain vulnerable to adversarial attacks, where imperceptible input perturbations fool the network. We present both theoretical and empirical analyses that connect the adversarial robustness of a model to the number of tasks that it is trained on. Experiments on two datasets show that attack difficulty increases as the number of target tasks increase. Moreover, our results suggest that when models are trained on multiple tasks at once, they become more robust to adversarial attacks on individual tasks. While adversarial defense remains an open challenge, our results suggest that deep networks are vulnerable partly because they are trained on too few tasks.

READ FULL TEXT

page 2

page 8

page 11

page 13

research
03/26/2021

Adversarial Attacks are Reversible with Natural Supervision

We find that images contain intrinsic structure that enables the reversa...
research
07/26/2019

Understanding Adversarial Robustness: The Trade-off between Minimum and Average Margin

Deep models, while being extremely versatile and accurate, are vulnerabl...
research
03/30/2023

Generating Adversarial Samples in Mini-Batches May Be Detrimental To Adversarial Robustness

Neural networks have been proven to be both highly effective within comp...
research
12/07/2019

Does Interpretability of Neural Networks Imply Adversarial Robustness?

The success of deep neural networks is clouded by two issues that largel...
research
04/07/2018

Fortified Networks: Improving the Robustness of Deep Networks by Modeling the Manifold of Hidden Representations

Deep networks have achieved impressive results across a variety of impor...
research
12/14/2019

Deep Poisoning Functions: Towards Robust Privacy-safe Image Data Sharing

As deep networks are applied to an ever-expanding set of computer vision...
research
09/11/2019

Structural Robustness for Deep Learning Architectures

Deep Networks have been shown to provide state-of-the-art performance in...

Please sign up or login with your details

Forgot password? Click here to reset