Disrupting adversarial transferability in deep neural networks

Wiedeman, Christopher; Wang, Ge

doi:10.1016/j.patter.2022.100472

Cited by 4 publications

(2 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the one hand, reducing the strength of the model's loss field can achieve the effect of enhancing the model's smoothness [75]. On the other hand, boosting the diversity of gradient orthogonality and reducing the magnitude of the gradient can also constrain the adversarial transferability [40,80], which can be explained as a special case of loss field orthogonality when the number of sampling points is m = 0.…”

Section: T ≤mentioning

confidence: 99%

Adversarial Robust Aerial Image Recognition Based on Reactive-Proactive Defense Framework with Deep Ensembles

Lu,

Sun,

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

As a safety-related application, visual systems based on deep neural networks (DNNs) in modern unmanned aerial vehicles (UAVs) show adversarial vulnerability when performing real-time inference. Recently, deep ensembles with various defensive strategies against adversarial samples have drawn much attention due to the increased diversity and reduced variance for their members. Aimed at the recognition task of remote sensing images (RSIs), this paper proposes to use a reactive-proactive ensemble defense framework to solve the security problem. In reactive defense, we fuse scoring functions of several classical detection algorithms with the hidden features and average output confidences from sub-models as a second fusion. In terms of proactive defense, we attempt two strategies, including enhancing the robustness of each sub-model and limiting the transferability among sub-models. In practical applications, the real-time RSIs are first input to the reactive defense part, which can detect and reject the adversarial RSIs. The accepted ones are then passed to robust recognition with a proactive defense. We conduct extensive experiments on three benchmark RSI datasets (i.e., UCM, AID, and FGSC-23). The experimental results show that the deep ensemble method of reactive and proactive defense performs very well in gradient-based attacks. The analysis of the applicable attack scenarios for each proactive ensemble defense is also helpful for this field. We also perform a case study with the whole framework in the black-box scenario, and the highest detection rate reaches 93.25%. Most of the adversarial RSIs can be rejected in advance or correctly recognized by the enhanced deep ensemble. This article is the first one to combine reactive and proactive defenses with a deep ensemble against adversarial attacks in the context of RSI recognition for DNN-based UAVs.

show abstract

Section: T ≤mentioning

confidence: 99%

Adversarial Robust Aerial Image Recognition Based on Reactive-Proactive Defense Framework with Deep Ensembles

Lu,

Sun,

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…[10]. In Reference [50], the authors proposed that transferability between seemingly different models is due to a high linear correlation between the feature sets extracted by different networks. In Reference [51], a systematic investigation of factors affecting adversarial examples' transferability for text classification models was explored.…”

mentioning

confidence: 99%

A Survey of Adversarial Attacks: An Open Issue for Deep Learning Sentiment Analysis Models

Vázquez-Hernández,

Morales-Rosales,

Algredo-Badillo

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

In recent years, the use of deep learning models for deploying sentiment analysis systems has become a widespread topic due to their processing capacity and superior results on large volumes of information. However, after several years’ research, previous works have demonstrated that deep learning models are vulnerable to strategically modified inputs called adversarial examples. Adversarial examples are generated by performing perturbations on data input that are imperceptible to humans but that can fool deep learning models’ understanding of the inputs and lead to false predictions being generated. In this work, we collect, select, summarize, discuss, and comprehensively analyze research works to generate textual adversarial examples. There are already a number of reviews in the existing literature concerning attacks on deep learning models for text applications; in contrast to previous works, however, we review works mainly oriented to sentiment analysis tasks. Further, we cover the related information concerning generation of adversarial examples to make this work self-contained. Finally, we draw on the reviewed literature to discuss adversarial example design in the context of sentiment analysis tasks.

show abstract

Disparity Analysis Between the Assembly and Byte Malware Samples with Deep Autoencoders

Abdullah

Cai

et al. 2022

2022 19th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)

View full text Add to dashboard Cite

Disrupting adversarial transferability in deep neural networks

Cited by 4 publications

References 11 publications

Adversarial Robust Aerial Image Recognition Based on Reactive-Proactive Defense Framework with Deep Ensembles

Adversarial Robust Aerial Image Recognition Based on Reactive-Proactive Defense Framework with Deep Ensembles

A Survey of Adversarial Attacks: An Open Issue for Deep Learning Sentiment Analysis Models

Disparity Analysis Between the Assembly and Byte Malware Samples with Deep Autoencoders

Contact Info

Product

Resources

About