Exploiting epistemic uncertainty of the deep learning models to generate adversarial samples
MetadataShow full item record
CitationTuna, Ö. F., Çatak, F. Ö. & Eskil, M. T. (2022). Exploiting epistemic uncertainty of the deep learning models to generate adversarial samples. Multimedia Tools and Applications, 1-22. doi:10.1007/s11042-022-12132-7
Deep neural network (DNN) architectures are considered to be robust to random perturbations. Nevertheless, it was shown that they could be severely vulnerable to slight but carefully crafted perturbations of the input, termed as adversarial samples. In recent years, numerous studies have been conducted in this new area called ``Adversarial Machine Learning” to devise new adversarial attacks and to defend against these attacks with more robust DNN architectures. However, most of the current research has concentrated on utilising model loss function to craft adversarial examples or to create robust models. This study explores the usage of quantified epistemic uncertainty obtained from Monte-Carlo Dropout Sampling for adversarial attack purposes by which we perturb the input to the shifted-domain regions where the model has not been trained on. We proposed new attack ideas by exploiting the difficulty of the target model to discriminate between samples drawn from original and shifted versions of the training data distribution by utilizing epistemic uncertainty of the model. Our results show that our proposed hybrid attack approach increases the attack success rates from 82.59% to 85.14%, 82.96% to 90.13% and 89.44% to 91.06% on MNIST Digit, MNIST Fashion and CIFAR-10 datasets, respectively.
SourceMultimedia Tools and Applications
The following license files are associated with this item:
Showing items related by title, author, creator and subject.
Extension of conventional co-training learning strategies to three-view and committee-based learning strategies for effective automatic sentence segmentation The objective of this work is to develop effective multi-view semi-supervised machine learning strategies for sentence boundary classification problem when only small sets of sentence boundary labeled data are available. ...
The primary objective of sentence segmentation process is to determine the sentence boundaries of a stream of words output by the automatic speech recognizers. Statistical methods developed for sentence segmentation requires ...
Yıldız, Olcay Taner; Bayrak, Ahmet Tuğrul; Özbek, Eyüp Erkan; Kestepe, Sedat (Institute of Electrical and Electronics Engineers Inc., 2019-09)Otel sayısının her geçen gün arttığı turizm sektöründe, aracı firmaların tüm oteller ile ayrı ayrı çalışma imkanı bulunmadığından, firmalar dünya üzerinde bir çok otelle anlaşması bulunan servis sağlayıcılarıyla beraber ...