141 citations · 766 across the 81 of their papers we have counts for
8 papers · 1 filter
Revisiting Outer Optimization in Adversarial Training
Ali Dabouei, Fariborz Taherkhani, Sobhan Soleymani +1
Despite the fundamental distinction between adversarial and natural training (AT and NT), AT methods generally adopt momentum SGD (MSGD) for the outer optimization. This paper aims…
SmoothFool: An Efficient Framework for Computing Smooth Adversarial Perturbations
Ali Dabouei, Sobhan Soleymani, Fariborz Taherkhani +2
Deep neural networks are susceptible to adversarial manipulations in the input domain. The extent of vulnerability has been explored intensively in cases of -bounded and $\…
Adversarial Examples to Fool Iris Recognition Systems
Sobhan Soleymani, Ali Dabouei, Jeremy Dawson +1
Adversarial examples have recently proven to be able to fool deep learning methods by adding carefully crafted small perturbation to the input space image. In this paper, we study…
Fast Geometrically-Perturbed Adversarial Faces
Ali Dabouei, Sobhan Soleymani, Jeremy Dawson +1
The state-of-the-art performance of deep learning algorithms has led to a considerable increase in the utilization of machine learning in security-sensitive and critical applicatio…
Multi-Level Feature Abstraction from Convolutional Neural Networks for Multimodal Biometric Identification
Sobhan Soleymani, Ali Dabouei, Hadi Kazemi +2
In this paper, we propose a deep multimodal fusion network to fuse multiple modalities (face, iris, and fingerprint) for person identification. The proposed deep multimodal fusion…
Generalized Bilinear Deep Convolutional Neural Networks for Multimodal Biometric Identification
Sobhan Soleymani, Amirsina Torfi, Jeremy Dawson +1
In this paper, we propose to employ a bank of modality-dedicated Convolutional Neural Networks (CNNs), fuse, train, and optimize them together for person classification tasks. A mo…
Attention-Based Guided Structured Sparsity of Deep Neural Networks
Amirsina Torfi, Rouzbeh A. Shirvani, Sobhan Soleymani +1
Network pruning is aimed at imposing sparsity in a neural network architecture by increasing the portion of zero-valued weights for reducing its size regarding energy-efficiency co…
Learning to classify with possible sensor failures
Tianpei Xie, Nasser M. Nasrabadi, Alfred O. Hero
In this paper, we propose a general framework to learn a robust large-margin binary classifier when corrupt measurements, called anomalies, caused by sensor failure might be presen…