Log In Sign Up

Output Randomization: A Novel Defense for both White-box and Black-box Adversarial Models

by   Daniel Park, et al.

Adversarial examples pose a threat to deep neural network models in a variety of scenarios, from settings where the adversary has complete knowledge of the model in a "white box" setting and to the opposite in a "black box" setting. In this paper, we explore the use of output randomization as a defense against attacks in both the black box and white box models and propose two defenses. In the first defense, we propose output randomization at test time to thwart finite difference attacks in black box settings. Since this type of attack relies on repeated queries to the model to estimate gradients, we investigate the use of randomization to thwart such adversaries from successfully creating adversarial examples. We empirically show that this defense can limit the success rate of a black box adversary using the Zeroth Order Optimization attack to 0 against white box adversaries. Unlike prior approaches that use randomization, our defense does not require its use at test time, eliminating the Backward Pass Differentiable Approximation attack, which was shown to be effective against other randomization defenses. Additionally, this defense has low overhead and is easily implemented, allowing it to be used together with other defenses across various model architectures. We evaluate output randomization training against the Projected Gradient Descent attacker and show that the defense can reduce the PGD attack's success rate down to 12 cross-entropy loss.


page 1

page 2

page 3

page 4


Thwarting finite difference adversarial attacks with output randomization

Adversarial examples pose a threat to deep neural network models in a va...

Adversarial Defense via Image Denoising with Chaotic Encryption

In the literature on adversarial examples, white box and black box attac...

Adversarial Robustness by Design through Analog Computing and Synthetic Gradients

We propose a new defense mechanism against adversarial attacks inspired ...

Defending against Adversarial Images using Basis Functions Transformations

We study the effectiveness of various approaches that defend against adv...

Adversarially Robust Classification by Conditional Generative Model Inversion

Most adversarial attack defense methods rely on obfuscating gradients. T...

An Adaptive Black-box Defense against Trojan Attacks (TrojDef)

Trojan backdoor is a poisoning attack against Neural Network (NN) classi...