GDGhislain Durif

Ghislain Durif

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Ghislain Durif has a total of 16 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

    • 11 May 2022
    • Published by HAL in HAL
    Open Access

    A median test for functional data

    by Zaineb Smida, Lionel Cucala, Ali Gannoun, Ghislain Durif

    The median test is more powerful than the Student t-test and the Wilcoxon-Mann-Whitney test in heavy-tailed cases for univariate data. When dealing with multidimensional data, the multivariate extension of the median test, called the sign test, is more efficient than the Hotelling T 2 and the Wilcoxon-Mann-Whitney tests for high dimensions and in very heavy-tailed cases. In this paper, we construct a median type test based on spatial ranks for functional data, i.e in infinite dimensional space, and we obtain asymptotic results. Then, we compare the proposed functional median test with the existing ones using simulated and real functional data.

  • Open Access

    Kernel Operations on the GPU, with Autodiff, without Memory Overflows

    by Benjamin Charlier, Jean Feydy, Joan Alexis Glaunès, François-David Collin, Ghislain Durif

    The KeOps library provides a fast and memory-efficient GPU support for tensors whose entries are given by a mathematical formula, such as kernel and distance matrices. KeOps alleviates the major bottleneck of tensor-centric libraries for kernel and geometric applications: memory consumption. It also supports automatic differentiation and outperforms standard GPU baselines, including PyTorch CUDA tensors or the Halide and TVM libraries. KeOps combines optimized C++/CUDA schemes with binders for high-level languages: Python (Numpy and PyTorch), Matlab and GNU R. As a result, high-level "quadratic" codes can now scale up to large data sets with millions of samples processed in seconds. KeOps brings graphics-like performances for kernel methods and is freely available on standard repositories (PyPi, CRAN). To showcase its versatility, we provide tutorials in a wide range of settings online at \url{www.kernel-operations.io}.


    • 24 March 2020
    • Published by HAL in HAL
    Open Access

    Kernel Operations on GPU, without memory overflows Kernel Operations on the GPU, with Autodiff, without Memory Overflows

    by Benjamin Charlier, Jean Feydy, Joan Glaunès, François-David Collin, Ghislain Durif

    The KeOps library provides a fast and memory-efficient GPU support for tensors whose entries are given by a mathematical formula, such as kernel and distance matrices. KeOps alleviates the major bottleneck of tensor-centric libraries for kernel and geometric applications: memory consumption. It also supports automatic differentiation and outperforms standard GPU baselines, including PyTorch CUDA tensors or the Halide and TVM libraries. KeOps combines optimized C++/CUDA schemes with binders for high-level languages: Python (Numpy and PyTorch), Matlab and GNU R. As a result, high-level "quadratic" codes can now scale up to large data sets with millions of samples processed in seconds. KeOps brings graphics-like performances for kernel methods and is freely available on standard repositories (PyPi, CRAN). To showcase its versatility, we provide tutorials in a wide range of settings online at www.kernel-operations.io.

  • Open Access

    Probabilistic count matrix factorization for single cell expression data analysis

    by Ghislain Durif, Laurent Modolo, Jeff E Mold, Sophie Lambert-Lacroix, Franck Picard

    The development of high-throughput single-cell sequencing technologies now allows the investigation of the population diversity of cellular transcriptomes. The expression dynamics (gene-to-gene variability) can be quantified more accurately, thanks to the measurement of lowly expressed genes. In addition, the cell-to-cell variability is high, with a low proportion of cells expressing the same genes at the same time/level. Those emerging patterns appear to be very challenging from the statistical point of view, especially to represent a summarized view of single-cell expression data. Principal component analysis (PCA) is a most powerful tool for high dimensional data representation, by searching for latent directions catching the most variability in the data. Unfortunately, classical PCA is based on Euclidean distance and projections that poorly work in presence of over-dispersed count data with dropout events like single-cell expression data. We propose a probabilistic Count Matrix Factorization (pCMF) approach for single-cell expression data analysis that relies on a sparse Gamma-Poisson factor model. This hierarchical model is inferred using a variational EM algorithm. It is able to jointly build a low dimensional representation of cells and genes. We show how this probabilistic framework induces a geometry that is suitable for single-cell data visualization, and produces a compression of the data that is very powerful for clustering purposes. Our method is competed against other standard representation methods like t-SNE, and we illustrate its performance for the representation of single-cell expression data. Our work is implemented in the pCMF R-package (https://github.com/gdurif/pCMF). Supplementary data are available at Bioinformatics online.


  • Open Access

    Probabilistic Count Matrix Factorization for Single Cell Expression Data Analysis

    by Ghislain Durif, Laurent Modolo, Jeff E. Mold, Sophie Lambert-Lacroix, Franck Picard

    The development of high throughput single-cell sequencing technologies now allows the investigation of the population level diversity of cellular transcriptomes. This diversity has shown two faces. First, the expression dynamics (gene to gene variability) can be quantified more accurately, thanks to the measurement of lowly-expressed genes. Second, the cell-to-cell variability is high, with a low proportion of cells expressing the same gene at the same time/level. Those emerging patterns appear to be very challenging from the statistical point of view, especially to represent and to provide a summarized view of single-cell expression data. PCA is one of the most powerful framework to provide a suitable representation of high dimensional datasets, by searching for latent directions catching the most variability in the data. Unfortunately, classical PCA is based on Euclidean distances and projections that work poorly in presence of over-dispersed counts that show drop-out events (zero-inflation) like single-cell expression data We propose a probabilistic Count Matrix Factorization (pCMF) approach for single-cell expression data analysis, that relies on a sparse Gamma-Poisson factor model. This hierarchical model is inferred using a variational EM algorithm. It is able to jointly build a low dimensional representation of cells and genes. We show how this probabilistic framework induces a geometry that is suitable for single-cell data visualization, and produces a compression of the data that is very powerful for clustering purposes. Our method is competed against other standard representation methods like t-SNE, and we illustrate its performance for the representation of single-cell data. We especially focus on publicly available single-cell RNA-seq datasets.

    • 31 January 2019
    • Published by HAL in HAL
    Open Access

    Probabilistic Count Matrix Factorization for Single Cell Expression Data Analysis

    by G Durif, L Modolo, J Mold, S Lambert-Lacroix, F Picard

    Motivation: The development of high throughput single-cell sequencing technologies now allows the investigation of the population diversity of cellular transcriptomes. The expression dynamics (gene-to-gene variability) can be quantified more accurately, thanks to the measurement of lowly-expressed genes. In addition, the cell-to-cell variability is high, with a low proportion of cells expressing the same genes at the same time/level. Those emerging patterns appear to be very challenging from the statistical point of view, especially to represent a summarized view of single-cell expression data. PCA is a most powerful tool for high dimensional data representation, by searching for latent directions catching the most variability in the data. Unfortunately, classical PCA is based on Euclidean distance and projections that poorly work in presence of over-dispersed count data with dropout events like single-cell expression data.Results: We propose a probabilistic Count Matrix Factorization (pCMF) approach for single-cell expression data analysis, that relies on a sparse Gamma-Poisson factor model. This hierarchical model is inferred using a variational EM algorithm. It is able to jointly build a low dimensional representation of cells and genes. We show how this probabilistic framework induces a geometry that is suitable for single-cell data visualization, and produces a compression of the data that is very powerful for clustering purposes. Our method is competed against other standard representation methods like t-SNE, and we illustrate its performance for the representation of single-cell expression (scRNA-seq) data.Availability: Our work is implemented in the pCMF R-package1.