11 sonuçlar
Arama Sonuçları
Listeleniyor 1 - 10 / 11
Yayın Convolutional attention network for MRI-based Alzheimer's disease classification and its interpretability analysis(IEEE, 2021-09-17) Türkan, Yasemin; Tek, Faik BorayNeuroimaging techniques, such as Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET), help to identify Alzheimer's disease (AD). These techniques generate large-scale, high-dimensional, multimodal neuroimaging data, which is time-consuming and difficult to interpret and classify. Therefore, interest in deep learning approaches for the classification of 3D structural MRI brain scans has grown rapidly. In this research study, we improved the 3D VGG model proposed by Korolev et al. [2]. We increased the filters in the 3D convolutional layers and then added an attention mechanism for better classification. We compared the performance of the proposed approaches for the classification of Alzheimer's disease versus mild cognitive impairments and normal cohorts on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset. We observed that both the accuracy and area under curve results improved with the proposed models. However, deep neural networks are black boxes that produce predictions that require further explanation for medical usage. We compared the 3D-data interpretation capabilities of the proposed models using four different interpretability methods: Occlusion, 3D Ultrametric Contour Map, 3D Gradient-Weighted Class Activation Mapping, and SHapley Additive explanations (SHAP). We observed that explanation results differed in different network models and data classes.Yayın Adaptive convolution kernel for artificial neural networks(Academic Press Inc., 2021-02) Tek, Faik Boray; Çam, İlker; Karlı, DenizMany deep neural networks are built by using stacked convolutional layers of fixed and single size (often 3 × 3) kernels. This paper describes a method for learning the size of convolutional kernels to provide varying size kernels in a single layer. The method utilizes a differentiable, and therefore backpropagation-trainable Gaussian envelope which can grow or shrink in a base grid. Our experiments compared the proposed adaptive layers to ordinary convolution layers in a simple two-layer network, a deeper residual network, and a U-Net architecture. The results in the popular image classification datasets such as MNIST, MNIST-CLUTTERED, CIFAR-10, Fashion, and ‘‘Faces in the Wild’’ showed that the adaptive kernels can provide statistically significant improvements on ordinary convolution kernels. A segmentation experiment in the Oxford-Pets dataset demonstrated that replacing ordinary convolution layers in a U-shaped network with 7 × 7 adaptive layers can improve its learning performance and ability to generalize.Yayın Automatic propbank generation for Turkish(Incoma Ltd, 2019-09) Ak, Koray; Yıldız, Olcay TanerSemantic role labeling (SRL) is an important task for understanding natural languages, where the objective is to analyse propositions expressed by the verb and to identify each word that bears a semantic role. It provides an extensive dataset to enhance NLP applications such as information retrieval, machine translation, information extraction, and question answering. However, creating SRL models are difficult. Even in some languages, it is infeasible to create SRL models that have predicate-argument structure due to lack of linguistic resources. In this paper, we present our method to create an automatic Turkish PropBank by exploiting parallel data from the translated sentences of English PropBank. Experiments show that our method gives promising results. © 2019 Association for Computational Linguistics (ACL).Yayın Malaria parasite detection with deep transfer learning(IEEE, 2018-12-06) Var, Esra; Tek, Faik BorayThis study aims to automatically detect malaria parasites (Plasmodium sp) on images taken from Giemsa stained blood smears. Deep learning methods provide limited performance when sample size is low. In transfer learning, visual features are learned from large general data sets, and problem-specific classification problem can be solved successfully in restricted problem specific data sets. In this study, we apply transfer learning method to detect and classify malaria parasites. We use a popular pre-trained CNN model VGG19. We trained the model for 20 epoch on 1428 P Vivax, 1425 P Ovule, 1446 E Falciparum, 1450 P Malariae and 1440 non-parasite samples. The transfer learning model achieves %80, %83, %86, %75 precision and 83%, 86%, 86%, 79% f-measure on 19 test images.Yayın Closeness and uncertainty aware adversarial examples detection in adversarial machine learning(Elsevier Ltd, 2022-07) Tuna, Ömer Faruk; Çatak, Ferhat Özgür; Eskil, Mustafa TanerWhile deep learning models are thought to be resistant to random perturbations, it has been demonstrated that these architectures are vulnerable to deliberately crafted perturbations, albeit being quasi-imperceptible. These vulnerabilities make it challenging to deploy Deep Neural Network (DNN) models in security-critical areas. Recently, many research studies have been conducted to develop defense techniques enabling more robust models. In this paper, we target detecting adversarial samples by differentiating them from their clean equivalents. We investigate various metrics for detecting adversarial samples. We first leverage moment-based predictive uncertainty estimates of DNN classifiers derived through Monte-Carlo (MC) Dropout Sampling. We also introduce a new method that operates in the subspace of deep features obtained by the model. We verified the effectiveness of our approach on different datasets. Our experiments show that these approaches complement each other, and combined usage of all metrics yields 99 % ROC-AUC adversarial detection score for well-known attack algorithms.Yayın An open, extendible, and fast Turkish morphological analyzer(Incoma Ltd, 2019-09) Yıldız, Olcay Taner; Avar, Begüm; Ercan, GökhanIn this paper, we present a two-level morphological analyzer for Turkish which consists of five main components: finite state transducer, rule engine for suffixation, lexicon, trie data structure, and LRU cache. We use Java language to implement finite state machine logic and rule engine, Xml language to describe the finite state transducer rules of the Turkish language, which makes the morphological analyzer both easily extendible and easily applicable to other languages. Empowered with a comprehensive lexicon of 54,000 bare-forms including 19,000 proper nouns, our morphological analyzer is amongst the most reliable analyzers produced so far. The analyzer is compared with Turkish morphological analyzers in the literature. By using LRU cache and a trie data structure, the system can analyze 100,000 words per second, which enables users to analyze huge corpora in a few hours.Yayın Animal sound classification using a convolutional neural network(IEEE, 2018-12-06) Şaşmaz, Emre; Tek, Faik BorayIn this paper, we investigate the problem of animal sound classification using deep learning and propose a system based on convolutional neural network architecture. As the input to the network, sound files were preprocessed to extract Mel Frequency Cepstral Coefficients (MFCC) using LibROSA library. To train and test the system we have collected 875 animal sound samples from an online sound source site for 10 different animal types. We report classification confusion matrices and the results obtained by different gradient descent optimizers. The best accuracy of 75% was obtained by Nesterov-accelerated Adaptive Moment Estimation (Nadam).Yayın Uncertainty as a Swiss army knife: new adversarial attack and defense ideas based on epistemic uncertainty(Springer, 2022-04-02) Tuna, Ömer Faruk; Çatak, Ferhat Özgür; Eskil, Mustafa TanerAlthough state-of-the-art deep neural network models are known to be robust to random perturbations, it was verified that these architectures are indeed quite vulnerable to deliberately crafted perturbations, albeit being quasi-imperceptible. These vulnerabilities make it challenging to deploy deep neural network models in the areas where security is a critical concern. In recent years, many research studies have been conducted to develop new attack methods and come up with new defense techniques that enable more robust and reliable models. In this study, we use the quantified epistemic uncertainty obtained from the model's final probability outputs, along with the model's own loss function, to generate more effective adversarial samples. And we propose a novel defense approach against attacks like Deepfool which result in adversarial samples located near the model's decision boundary. We have verified the effectiveness of our attack method on MNIST (Digit), MNIST (Fashion) and CIFAR-10 datasets. In our experiments, we showed that our proposed uncertainty-based reversal method achieved a worst case success rate of around 95% without compromising clean accuracy.Yayın Exploiting epistemic uncertainty of the deep learning models to generate adversarial samples(Springer, 2022-03) Tuna, Ömer Faruk; Çatak, Ferhat Özgür; Eskil, Mustafa TanerDeep 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.Yayın BOUN-ISIK participation: an unsupervised approach for the named entity normalization and relation extraction of Bacteria Biotopes(Association for Computational Linguistics (ACL), 2019-11-04) Karadeniz, İlknur; Tuna, Ömer Faruk; Özgu, ArzucanThis paper presents our participation at the Bacteria Biotope Task of the BioNLP Shared Task 2019. Our participation includes two systems for the two subtasks of the Bacteria Biotope Task: the normalization of entities (BB-norm) and the identification of the relations between the entities given a biomedical text (BB-rel). For the normalization of entities, we utilized word embeddings and syntactic re-ranking. For the relation extraction task, pre-defined rules are used. Although both approaches are unsupervised, in the sense that they do not need any labeled data, they achieved promising results. Especially, for the BB-norm task, the results have shown that the proposed method performs as good as deep learning based methods, which require labeled data.
