Publications by Alhussein Fawzi

Robustness via curvature regularization, and vice versa

Pascal Frossard, Alhussein Fawzi, Seyed Mohsen Moosavi Dezfooli

State-of-the-art classifiers have been shown to be largely vulnerable to adversarial perturbations. One of the most effective strategies to improve robustness is adversarial training. In this paper, we investigate the effect of adversarial training on the ...

IEEE2019

Analysis of classifiers' robustness to adversarial perturbations

Pascal Frossard, Alhussein Fawzi

2018

Robustness of Classifiers to Universal Perturbations: A Geometric Perspective

Pascal Frossard, Alhussein Fawzi, Seyed Mohsen Moosavi Dezfooli

Deep networks have recently been shown to be vulnerable to universal perturbations: there exist very small image-agnostic perturbations that cause most natural images to be misclassified by such classifiers. In this paper, we provide a quantitative analysi ...

2018

Empirical study of the topology and geometry of deep networks

Pascal Frossard, Alhussein Fawzi, Seyed Mohsen Moosavi Dezfooli

The goal of this paper is to analyze the geometric properties of deep neural network image classifiers in the input space. We specifically study the topology of classification regions created by deep networks, as well as their associated decision boundary. ...

2018

Universal adversarial perturbations

Pascal Frossard, Alhussein Fawzi, Seyed Mohsen Moosavi Dezfooli

Given a state-of-the-art deep neural network classifier, we show the existence of a universal (image-agnostic) and very small perturbation vector that causes natural images to be misclassified with high probability. We propose a systematic algorithm for co ...

Ieee2017

Multi-task additive models with shared transfer functions based on dictionary learning

Pascal Frossard, Alhussein Fawzi

Additive models form a widely popular class of regression models which represent the relation between covariates and response variables as the sum of low-dimensional transfer functions. Besides flexibility and accuracy, a key benefit of these models is the ...

Institute of Electrical and Electronics Engineers2017

The Robustness of Deep Networks - A geometric perspective

Pascal Frossard, Alhussein Fawzi, Seyed Mohsen Moosavi Dezfooli

Deep neural networks have recently shown impressive classification performance on a diverse set of visual tasks. When deployed in real-world (noise-prone) environments, it is equally important that these classifiers satisfy robustness guarantees: small per ...

Institute of Electrical and Electronics Engineers2017

Image Inpainting through Neural Networks Hallucinations

Pascal Frossard, Alhussein Fawzi

We consider in this paper the problem of image inpainting, where the objective is to reconstruct large continuous regions of missing or deteriorated parts of an image. Traditional inpainting algorithms are unfortunately not well adapted to handle such corr ...

Ieee2016

Robust image classification : analysis and applications

Alhussein Fawzi

In the past decade, image classification systems have witnessed major advances that led to record performances on challenging datasets. However, little is known about the behavior of these classifiers when the data is subject to perturbations, such as rand ...

EPFL2016

Structured Dimensionality Reduction for Additive Model Regression

Pascal Frossard, Alhussein Fawzi

Additive models are regression methods which model the response variable as the sum of univariate transfer functions of the input variables. Key benefits of additive models are their accuracy and interpretability on many real-world tasks. Additive models a ...

Institute of Electrical and Electronics Engineers2016

Measuring the effect of nuisance variables on classifiers

Pascal Frossard, Alhussein Fawzi

In real-world classification problems, nuisance variables can cause wild variability in the data. Nuisance corresponds for example to geometric distortions of the image, occlusions, illumination changes or any other deformations that do not alter the groun ...

2016

DeepFool: a simple and accurate method to fool deep neural networks

Pascal Frossard, Alhussein Fawzi, Seyed Mohsen Moosavi Dezfooli

State-of-the-art deep neural networks have achieved impressive results on many image classification tasks. However, these same architectures have been shown to be unstable to small, well sought, perturbations of the images. Despite the importance of this p ...

2016

Robustness of classifiers: from adversarial to random noise

Pascal Frossard, Alhussein Fawzi, Seyed Mohsen Moosavi Dezfooli

Several recent works have shown that state-of-the-art classifiers are vulnerable to worst-case (i.e., adversarial) perturbations of the datapoints. On the other hand, it has been empirically observed that these same classifiers are relatively robust to ran ...

2016

Adaptive data augmentation for image classification

Pascal Frossard, Alhussein Fawzi

Data augmentation is the process of generating samples by transforming training data, with the target of improving the accuracy and robustness of classifiers. In this paper, we propose a new automatic and adaptive algorithm for choosing the transformations ...

IEEE2016

Multi-task additive models with shared transfer functions based on dictionary learning

Pascal Frossard, Alhussein Fawzi

2015

Fundamental limits on adversarial robustness

Pascal Frossard, Alhussein Fawzi

2015

Dictionary learning for fast classification based on soft-thresholding

Pascal Frossard, Alhussein Fawzi

Classifiers based on sparse representations have recently been shown to provide excellent results in many visual recognition and classification tasks. However, the high cost of computing sparse representations at test time is a major obstacle that limits t ...

Springer2015

Manitest: Are classifiers really invariant?

Pascal Frossard, Alhussein Fawzi

Invariance to geometric transformations is a highly desirable property of automatic classifiers in many image recognition tasks. Nevertheless, it is unclear to which extent state-of-the-art classifiers are invariant to basic transformations such as rotatio ...

2015

Dictionary Learning for Efficient Classification based on Soft-Thresholding

Pascal Frossard, Alhussein Fawzi

2014

Image registration with sparse approximations in parametric dictionaries

Pascal Frossard, Alhussein Fawzi

We examine in this paper the problem of image registration from the new perspective where images are given by sparse approximations in parametric dictionaries of geometric functions. We propose a registration algorithm that looks for an estimate of the glo ...

2013