Rapid Plug-in Defenders

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wu, Kai, Li, Yujian Betterest, Lou, Jian, Zhang, Xiaoyu, Wang, Handing, Liu, Jing
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910679100817408
author Wu, Kai
Li, Yujian Betterest
Lou, Jian
Zhang, Xiaoyu
Wang, Handing
Liu, Jing
author_facet Wu, Kai
Li, Yujian Betterest
Lou, Jian
Zhang, Xiaoyu
Wang, Handing
Liu, Jing
contents In the realm of daily services, the deployment of deep neural networks underscores the paramount importance of their reliability. However, the vulnerability of these networks to adversarial attacks, primarily evasion-based, poses a concerning threat to their functionality. Common methods for enhancing robustness involve heavy adversarial training or leveraging learned knowledge from clean data, both necessitating substantial computational resources. This inherent time-intensive nature severely limits the agility of large foundational models to swiftly counter adversarial perturbations. To address this challenge, this paper focuses on the Rapid Plug-in Defender (RaPiD) problem, aiming to rapidly counter adversarial perturbations without altering the deployed model. Drawing inspiration from the generalization and the universal computation ability of pre-trained transformer models, we propose a novel method termed CeTaD (Considering Pre-trained Transformers as Defenders) for RaPiD, optimized for efficient computation. CeTaD strategically fine-tunes the normalization layer parameters within the defender using a limited set of clean and adversarial examples. Our evaluation centers on assessing CeTaD's effectiveness, transferability, and the impact of different components in scenarios involving one-shot adversarial examples. The proposed method is capable of rapidly adapting to various attacks and different application scenarios without altering the target model and clean training data. We also explore the influence of varying training data conditions on CeTaD's performance. Notably, CeTaD exhibits adaptability across differentiable service models and proves the potential of continuous learning.
format Preprint
id arxiv_https___arxiv_org_abs_2306_01762
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Rapid Plug-in Defenders
Wu, Kai
Li, Yujian Betterest
Lou, Jian
Zhang, Xiaoyu
Wang, Handing
Liu, Jing
Cryptography and Security
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
In the realm of daily services, the deployment of deep neural networks underscores the paramount importance of their reliability. However, the vulnerability of these networks to adversarial attacks, primarily evasion-based, poses a concerning threat to their functionality. Common methods for enhancing robustness involve heavy adversarial training or leveraging learned knowledge from clean data, both necessitating substantial computational resources. This inherent time-intensive nature severely limits the agility of large foundational models to swiftly counter adversarial perturbations. To address this challenge, this paper focuses on the Rapid Plug-in Defender (RaPiD) problem, aiming to rapidly counter adversarial perturbations without altering the deployed model. Drawing inspiration from the generalization and the universal computation ability of pre-trained transformer models, we propose a novel method termed CeTaD (Considering Pre-trained Transformers as Defenders) for RaPiD, optimized for efficient computation. CeTaD strategically fine-tunes the normalization layer parameters within the defender using a limited set of clean and adversarial examples. Our evaluation centers on assessing CeTaD's effectiveness, transferability, and the impact of different components in scenarios involving one-shot adversarial examples. The proposed method is capable of rapidly adapting to various attacks and different application scenarios without altering the target model and clean training data. We also explore the influence of varying training data conditions on CeTaD's performance. Notably, CeTaD exhibits adaptability across differentiable service models and proves the potential of continuous learning.
title Rapid Plug-in Defenders
topic Cryptography and Security
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2306.01762