3 sonuçlar
Arama Sonuçları
Listeleniyor 1 - 3 / 3
Yayın Unreasonable effectiveness of last hidden layer activations for adversarial robustness(Institute of Electrical and Electronics Engineers Inc., 2022) Tuna, Ömer Faruk; Çatak, Ferhat Özgür; Eskil, Mustafa TanerIn standard Deep Neural Network (DNN) based classifiers, the general convention is to omit the activation function in the last (output) layer and directly apply the softmax function on the logits to get the probability scores of each class. In this type of architectures, the loss value of the classifier against any output class is directly proportional to the difference between the final probability score and the label value of the associated class. Standard White-box adversarial evasion attacks, whether targeted or untargeted, mainly try to exploit the gradient of the model loss function to craft adversarial samples and fool the model. In this study, we show both mathematically and experimentally that using some widely known activation functions in the output layer of the model with high temperature values has the effect of zeroing out the gradients for both targeted and untargeted attack cases, preventing attackers from exploiting the model's loss function to craft adversarial samples. We've experimentally verified the efficacy of our approach on MNIST (Digit), CIFAR10 datasets. Detailed experiments confirmed that our approach substantially improves robustness against gradient-based targeted and untargeted attack threats. And, we showed that the increased non-linearity at the output layer has some ad-ditional benefits against some other attack methods like Deepfool attack.Yayın TENET: a new hybrid network architecture for adversarial defense(Springer Science and Business Media Deutschland GmbH, 2023-08) Tuna, Ömer Faruk; Çatak, Ferhat Özgür; Eskil, Mustafa TanerDeep neural network (DNN) models are widely renowned for their resistance to random perturbations. However, researchers have found out that these models are indeed extremely vulnerable to deliberately crafted and seemingly imperceptible perturbations of the input, referred to as adversarial examples. Adversarial attacks have the potential to substantially compromise the security of DNN-powered systems and posing high risks especially in the areas where security is a top priority. Numerous studies have been conducted in recent years to defend against these attacks and to develop more robust architectures resistant to adversarial threats. In this study, we propose a new architecture and enhance a recently proposed technique by which we can restore adversarial samples back to their original class manifold. We leverage the use of several uncertainty metrics obtained from Monte Carlo dropout (MC Dropout) estimates of the model together with the model’s own loss function and combine them with the use of defensive distillation technique to defend against these attacks. We have experimentally evaluated and verified the efficacy of our approach on MNIST (Digit), MNIST (Fashion) and CIFAR10 datasets. In our experiments, we showed that our proposed method reduces the attack’s success rate lower than 5% without compromising clean accuracy.Yayın Segmentation based classification of retinal diseases in OCT images(Institute of Electrical and Electronics Engineers Inc., 2024) Eren, Öykü; Tek, Faik Boray; Turkan, YaseminVolumetric optical coherence tomography (OCT) scans offer detailed visualization of the retinal layers, where any deformation can indicate potential abnormalities. This study introduced a method for classifying ocular diseases in OCT images through transfer learning. Applying transfer learning from natural images to Optical Coherence Tomography (OCT) scans present challenges, particularly when target domain examples are limited. Our approach aimed to enhance OCT-based retinal disease classification by leveraging transfer learning more effectively. We hypothesize that providing an explicit layer structure can improve classification accuracy. Using the OCTA-500 dataset, we explored various configurations by segmenting the retinal layers and integrating these segmentations with OCT scans. By combining horizontal and vertical cross-sectional middle slices and their blendings with segmentation outputs, we achieved a classification a ccuracy of 91.47% and an Area Under the Curve (AUC) of 0.96, significantly outperforming the classification of OCT slice images.
