paper-with-me

Papers

Stochastic Activation Pruning for Robust Adversarial Defense

2018-03-05 · ICLR 2018 1 · Guneet S. Dhillon, Kamyar Azizzadenesheli, Zachary C. Lipton, Jeremy Bernstein, Jean Kossaifi, Aran Khanna, Anima Anandkumar

Neural networks are known to be vulnerable to adversarial examples. Carefully chosen perturbations to real images, while imperceptible to humans, induce misclassification and threaten the reliability of deep learning systems in the wild. To guard against adversarial examples, we take inspiration from game theory and cast the problem as a minimax zero-sum game between the adversary and the model. In general, for such games, the optimal strategy for both players requires a stochastic policy, also known as a mixed strategy. In this light, we propose Stochastic Activation Pruning (SAP), a mixed strategy for adversarial defense. SAP prunes a random subset of activations (preferentially pruning those with smaller magnitude) and scales up the survivors to compensate. We can apply SAP to pretrained networks, including adversarially trained models, without fine-tuning, providing robustness against adversarial examples. Experiments demonstrate that SAP confers robustness against attacks, increasing accuracy and preserving calibration.

📄 PDF Abstract BibTeX arXiv:1803.01442

Code (1)

Guneet-Dhillon/Stochastic-Activation-Pruning 공식 구현

Tasks

Adversarial Defense

Methods 이 논문이 사용한 방법론

Pruning 설명 없음

Similar Papers 제목 키워드 기반

Erratum Concerning the Obfuscated Gradients Attack on Stochastic Activation Pruning

2020-09-30 · Guneet S. Dhillon, Nicholas Carlini

Stochastic Activation Pruning (SAP) (Dhillon et al., 2018) is a defense to adversarial examples that was attacked and found to be broken by the "Obfuscated Gradients" paper (Athalye et al., 2018). We discover a flaw in t…

Block Switching: A Stochastic Approach for Deep Learning Security

2020-02-18 · Xiao Wang, Siyue Wang, Pin-Yu Chen, Xue Lin 외

Recent study of adversarial attacks has revealed the vulnerability of modern deep learning models. That is, subtly crafted perturbations of the input can make a trained network with high accuracy produce arbitrary incorr…

Deep Learning

Protecting Neural Networks with Hierarchical Random Switching: Towards Better Robustness-Accuracy Trade-off for Stochastic Defenses

2019-08-20 · Xiao Wang, Siyue Wang, Pin-Yu Chen, Yanzhi Wang 외

Despite achieving remarkable success in various domains, recent studies have uncovered the vulnerability of deep neural networks to adversarial perturbations, creating concerns on model generalizability and new threats s…

Adversarial Robustness

Simple and Effective Stochastic Neural Networks

2019-09-25 · Tianyuan Yu, Yongxin Yang, Da Li, Timothy Hospedales 외

Stochastic neural networks (SNNs) are currently topical, with several paradigms being actively investigated including dropout, Bayesian neural networks, variational information bottleneck (VIB) and noise regularized lear…

Adversarial AttackAdversarial Defense

Pixel Redrawn For A Robust Adversarial Defense

2019-05-01 · ICLR 2019 5 · Jiacang Ho, Dae-Ki Kang

Recently, an adversarial example becomes a serious problem to be aware of because it can fool trained neural networks easily. To prevent the issue, many researchers have proposed several defense techniques such as advers…

Adversarial Defense