35 citations · 163 across the 31 of their papers we have counts for
23 papers · 1 filter
Towards Evading the Limits of Randomized Smoothing: A Theoretical Analysis
Raphael Ettedgui, Alexandre Araujo, Rafael Pinot +2
Randomized smoothing is the dominant standard for provable defenses against adversarial examples. Nevertheless, this method has recently been proven to suffer from important inform…
Towards Consistency in Adversarial Classification
Laurent Meunier, Raphaël Ettedgui, Rafael Pinot +2
In this paper, we study the problem of consistency in the context of adversarial examples. Specifically, we tackle the following question: can surrogate losses still be used as a p…
Non parametric estimation of causal populations in a counterfactual scenario
Celine Beji, Florian Yger, Jamal Atif
In causality, estimating the effect of a treatment without confounding inference remains a major issue because requires to assess the outcome in both case with and without treatmen…
Online certification of preference-based fairness for personalized recommender systems
Virginie Do, Sam Corbett-Davies, Jamal Atif +1
Recommender systems are facing scrutiny because of their growing impact on the opportunities we have access to. Current audits for fairness are limited to coarse-grained parity ass…
On the robustness of randomized classifiers to adversarial examples
Rafael Pinot, Laurent Meunier, Florian Yger +3
This paper investigates the theory of robustness against adversarial attacks. We focus on randomized classifiers (\emph{i.e.} classifiers that output random variables) and provide…
AAMDRL: Augmented Asset Management with Deep Reinforcement Learning
Eric Benhamou, David Saltiel, Sandrine Ungari +2
Can an agent learn efficiently in a noisy and self adapting environment with sequential, non-stationary and non-homogeneous observations? Through trading bots, we illustrate how De…
On Lipschitz Regularization of Convolutional Layers using Toeplitz Matrix Theory
Alexandre Araujo, Benjamin Negrevergne, Yann Chevaleyre +1
This paper tackles the problem of Lipschitz regularization of Convolutional Neural Networks. Lipschitz regularity is now established as a key property of modern deep learning with…
Estimating Individual Treatment Effects through Causal Populations Identification
Céline Beji, Michaël Bon, Florian Yger +1
Estimating the Individual Treatment Effect from observational data, defined as the difference between outcomes with and without treatment or intervention, while observing just one…
Randomization matters. How to defend against strong adversarial attacks
Rafael Pinot, Raphael Ettedgui, Geovani Rizk +2
Is there a classifier that ensures optimal robustness against all adversarial attacks? This paper answers this question by adopting a game-theoretic point of view. We show that adv…
Yet another but more efficient black-box adversarial attack: tiling and evolution strategies
Laurent Meunier, Jamal Atif, Olivier Teytaud
We introduce a new black-box attack achieving state of the art performances. Our approach is based on a new objective function, borrowing ideas from -white box attacks…
A unified view on differential privacy and robustness to adversarial examples
Rafael Pinot, Florian Yger, Cédric Gouy-Pailler +1
This short note highlights some links between two lines of research within the emerging topic of trustworthy machine learning: differential privacy and robustness to adversarial ex…
NGO-GM: Natural Gradient Optimization for Graphical Models
Eric Benhamou, Jamal Atif, Rida Laraki +1
This paper deals with estimating model parameters in graphical models. We reformulate it as an information geometric optimization problem and introduce a natural gradient descent s…
Robust Neural Networks using Randomized Adversarial Training
Alexandre Araujo, Laurent Meunier, Rafael Pinot +1
This paper tackles the problem of defending a neural network against adversarial attacks crafted with different norms (in particular and bounded adversarial…
Theoretical evidence for adversarial robustness through randomization
Rafael Pinot, Laurent Meunier, Alexandre Araujo +4
This paper investigates the theory of robustness against adversarial attacks. It focuses on the family of randomization techniques that consist in injecting noise in the network at…
Understanding and Training Deep Diagonal Circulant Neural Networks
Alexandre Araujo, Benjamin Negrevergne, Yann Chevaleyre +1
In this paper, we study deep diagonal circulant neural networks, that is deep neural networks in which weight matrices are the product of diagonal and circulant ones. Besides makin…
A discrete version of CMA-ES
Eric Benhamou, Jamal Atif, Rida Laraki
Modern machine learning uses more and more advanced optimization techniques to find optimal hyper parameters. Whenever the objective function is non-convex, non continuous and with…
A new approach to learning in Dynamic Bayesian Networks (DBNs)
E. Benhamou, J. Atif, R. Laraki
In this paper, we revisit the parameter learning problem, namely the estimation of model parameters for Dynamic Bayesian Networks (DBNs). DBNs are directed graphical models of stoc…
On the Needs for Rotations in Hypercubic Quantization Hashing
Anne Morvan, Antoine Souloumiac, Krzysztof Choromanski +2
The aim of this paper is to endow the well-known family of hypercubic quantization hashing methods with theoretical guarantees. In hypercubic quantization, applying a suitable (ran…
Streaming Binary Sketching based on Subspace Tracking and Diagonal Uniformization
Anne Morvan, Antoine Souloumiac, Cédric Gouy-Pailler +1
In this paper, we address the problem of learning compact similarity-preserving embeddings for massive high-dimensional streams of data in order to perform efficient similarity sea…
Graph sketching-based Space-efficient Data Clustering
Anne Morvan, Krzysztof Choromanski, Cédric Gouy-Pailler +1
In this paper, we address the problem of recovering arbitrary-shaped data clusters from datasets while facing \emph{high space constraints}, as this is for instance the case in man…
Structured adaptive and random spinners for fast machine learning computations
Mariusz Bojarski, Anna Choromanska, Krzysztof Choromanski +6
We consider an efficient computational framework for speeding up several machine learning algorithms with almost no loss of accuracy. The proposed framework relies on projections v…
TripleSpin - a generic compact paradigm for fast machine learning computations
Krzysztof Choromanski, Francois Fagan, Cedric Gouy-Pailler +3
We present a generic compact computational framework relying on structured random matrices that can be applied to speed up several machine learning algorithms with almost no loss o…
Metrics for Multivariate Dictionaries
Sylvain Chevallier, Quentin Barthélemy, Jamal Atif
Overcomplete representations and dictionary learning algorithms kept attracting a growing interest in the machine learning community. This paper addresses the emerging problem of c…