arXiv subjects

Laurent Risser

Publications and source records attributed to Laurent Risser.

27 records · Page 2Linked to original sources

arXiv2020

A survey of bias in Machine Learning through the prism of Statistical Parity for the Adult Data Set

Applications based on Machine Learning models have now become an indispensable part of the everyday life and the professional world. A critical question then recently arised among the population: Do algorithmic decisions convey any type of discrimination against specific groups of population or minorities? In this paper, we show the importance of understanding how a bias can be introduced into automatic decisions. We first present a mathematical framework for the fair learning problem, specifically in the binary classification setting. We then propose to quantify the presence of bias by using the standard Disparate Impact index on the real and well-known Adult income data set. Finally, we check the performance of different approaches aiming to reduce the bias in binary classification outcomes. Importantly, we show that some intuitive methods are ineffective. This sheds light on the fact trying to make fair machine learning models may be a particularly challenging task, in particular when the training observations contain a bias.

stat.ML↗

arXiv2019

Distribution regression model with a Reproducing Kernel Hilbert Space approach

In this paper, we introduce a new distribution regression model for probability distributions. This model is based on a Reproducing Kernel Hilbert Space (RKHS) regression framework, where universal kernels are built using Wasserstein distances for distributions belonging to W 2 ($Ω$) and $Ω$ is a compact subspace of R. We prove the universal kernel property of such kernels and use this setting to perform regressions on functions. Different regression models are first compared with the proposed one on simulated functional data. We then apply our regression model to transient evoked otoascoutic emission (TEOAE) distribution responses and real predictors of the age.

math.ST↗

arXiv2018

COREclust: a new package for a robust and scalable analysis of complex data

In this paper, we present a new R package COREclust dedicated to the detection of representative variables in high dimensional spaces with a potentially limited number of observations. Variable sets detection is based on an original graph clustering strategy denoted CORE-clustering algorithm that detects CORE-clusters, i.e. variable sets having a user defined size range and in which each variable is very similar to at least another variable. Representative variables are then robustely estimate as the CORE-cluster centers. This strategy is entirely coded in C++ and wrapped by R using the Rcpp package. A particular effort has been dedicated to keep its algorithmic cost reasonable so that it can be used on large datasets. After motivating our work, we will explain the CORE-clustering algorithm as well as a greedy extension of this algorithm. We will then present how to use it and results obtained on synthetic and real data.

cs.MS↗

arXiv2018

Online Barycenter Estimation of Large Weighted Graphs

In this paper, we propose a new method to compute the barycenter of large weighted graphs endowed with probability measures on their nodes. We suppose that the edge weights are distances between the nodes and that the probability measure on the nodes is related to events observed there. For instance, a graph can represent a subway network: its edge weights are the distance between two stations, and the observed events at each node are the subway users getting in or leaving the subway network at this station. The probability measure on the nodes does not need to be explicitly known. Our strategy only uses observed node related events to give more or less emphasis to the different nodes. Furthermore, the barycenter estimation can be updated in real time with each new event. We propose a multiscale extension of \cite{arXiv:1605.04148} where the decribed strategy is valid only for medium-sized graphs due to memory costs. Our multiscale approach is inspired from the geometrical decomposition of the barycenter in a Euclidean space: we apply a heuristic \textit{divide et impera} strategy based on a preliminary clustering. Our strategy is finally assessed on road- and social-networks of up to $10^6$ nodes. We show that its results compare well with \cite{arXiv:1605.04148} in terms of accuracy and stability on small graphs, and that it can additionally be used on large graphs even on standard laptops.

math.PR↗

arXiv2017

Cytometry inference through adaptive atomic deconvolution

In this paper we consider a statistical estimation problem known as atomic deconvolution. Introduced in reliability, this model has a direct application when considering biological data produced by flow cytometers. In these experiments, biologists measure the fluorescence emission of treated cells and compare them with their natural emission to study the presence of specific molecules on the cells' surface. They observe a signal which is composed of a noise (the natural fluorescence) plus some additional signal related to the quantity of molecule present on the surface if any. From a statistical point of view, we aim at inferring the percentage of cells expressing the selected molecule and the probability distribution function associated with its fluorescence emission. We propose here an adap-tive estimation procedure based on a previous deconvolution procedure introduced by [vEGS08, GvES11]. For both estimating the mixing parameter and the mixing density automatically, we use the Lepskii method based on the optimal choice of a bandwidth using a bias-variance decomposition. We then derive some concentration inequalities for our estimators and obtain the convergence rates, that are shown to be minimax optimal (up to some log terms) in Sobolev classes. Finally, we apply our algorithm on simulated and real biological data.

math.ST↗

arXiv2016

How to compute the barycenter of a weighted graph

Discrete structures like graphs make it possible to naturally and flexibly model complex phenomena. Since graphs that represent various types of information are increasingly available today, their analysis has become a popular subject of research. The graphs studied in the field of data science at this time generally have a large number of nodes that are not fairly weighted and connected to each other, translating a structural specification of the data. Yet, even an algorithm for locating the average position in graphs is lacking although this knowledge would be of primary interest for statistical or representation problems. In this work, we develop a stochastic algorithm for finding the Frechet mean of weighted undirected metric graphs. This method relies on a noisy simulated annealing algorithm dealt with using homogenization. We then illustrate our algorithm with two examples (subgraphs of a social network and of a collaboration and citation network).

math.PR↗

arXiv2015

Mixture of Kernels and Iterated Semidirect Product of Diffeomorphisms Groups

In the framework of large deformation diffeomorphic metric mapping (LDDMM), we develop a multi-scale theory for the diffeomorphism group based on previous works. The purpose of the paper is (1) to develop in details a variational approach for multi-scale analysis of diffeomorphisms, (2) to generalise to several scales the semidirect product representation and (3) to illustrate the resulting diffeomorphic decomposition on synthetic and real images. We also show that the approaches presented in other papers and the mixture of kernels are equivalent.

math.NA↗

arXiv2015

Anisotropic Diffusion in ITK

Anisotropic Non-Linear Diffusion is a powerful image processing technique, which allows to simultaneously remove the noise and enhance sharp features in two or three dimensional images. Anisotropic Diffusion is understood here in the sense of Weickert, meaning that diffusion tensors are anisotropic and reflect the local orientation of image features. This is in contrast with the non-linear diffusion filter of Perona and Malik, which only involves scalar diffusion coefficients, in other words isotropic diffusion tensors. In this paper, we present an anisotropic non-linear diffusion technique we implemented in ITK. This technique is based on a recent adaptive scheme making the diffusion stable and requiring limited numerical resources. (See supplementary data.)

cs.CV↗

arXiv2014

Diffeomorphic image matching with left-invariant metrics

The geometric approach to diffeomorphic image registration known as "large deformation by diffeomorphic metric mapping" (LDDMM) is based on a left action of diffeomorphisms on images, and a right-invariant metric on a diffeomorphism group, usually defined using a reproducing kernel. We explore the use of left-invariant metrics on diffeomorphism groups, based on reproducing kernels defined in the body coordinates of a source image. This perspective, which we call Left-LDM, allows us to consider non-isotropic spatially-varying kernels, which can be interpreted as describing variable deformability of the source image. We also show a simple relationship between LDDMM and the new approach, implying that spatially-varying kernels are interpretable in the same way in LDDMM. We conclude with a discussion of a class of kernels that enforce a soft mirror-symmetry constraint, which we validate in numerical experiments on a model of a lesioned brain.

math.DG↗