Person
Michaël Defferrard
This person is no longer with EPFL
Related publications (9)
Leveraging topology, geometry, and symmetries for efficient Machine Learning
When learning from data, leveraging the symmetries of the domain the data lies on is a principled way to combat the curse of dimensionality: it constrains the set of functions to learn from. It is more data efficient than augmentation and gives a generaliz ...
EPFL2022Connectome spectral analysis to track EEG task dynamics on a subsecond scale
David Pascucci, Sébastien Tourbier, Patric Hagmann, Gijs Plomp, Michaël Defferrard
We present an approach for tracking fast spatiotemporal cortical dynamics in which we combine white matter connectivity data with source-projected electroencephalographic (EEG) data. We employ the mathematical framework of graph signal processing in order ...
2020DeepSphere: Efficient spherical convolutional neural network with HEALPix sampling for cosmological applications
Nathanaël Perraudin, Michaël Defferrard
Convolutional Neural Networks (CNNs) are a cornerstone of the Deep Learning toolbox and have led to many breakthroughs in Artificial Intelligence. So far, these neural networks (NNs) have mostly been developed for regular Euclidean domains such as those su ...
ELSEVIER SCIENCE BV2019Structured Sequence Modeling with Graph Convolutional Recurrent Networks
Pierre Vandergheynst, Xavier Bresson, Michaël Defferrard, Youngjoo Seo
This paper introduces Graph Convolutional Recurrent Network (GCRN), a deep learning model able to predict structured sequences of data. Precisely, GCRN is a generalization of classical recurrent neural networks (RNN) to data structured by an arbitrary grap ...
2017FMA: A Dataset For Music Analysis
Pierre Vandergheynst, Kirell Maël Benzi, Xavier Bresson, Michaël Defferrard
We introduce the Free Music Archive (FMA), an open and easily accessible dataset which can be used to evaluate several tasks in music information retrieval (MIR), a field concerned with browsing, searching, and organizing large music collections. The commu ...
2017Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering
Pierre Vandergheynst, Xavier Bresson, Michaël Defferrard
In this work, we are interested in generalizing convolutional neural networks (CNNs) from low-dimensional regular grids, where image, video and speech are represented, to high-dimensional irregular domains, such as social networks, brain connectomes or wor ...
2016Structured Auto-Encoder with application to Music Genre Recognition
In this work, we present a technique that learns discriminative audio features for Music Information Retrieval (MIR). The novelty of the proposed technique is to design auto-encoders that make use of data structures to learn enhanced sparse data representa ...
2015Graph-based Image Inpainting
The project goal was to explore the applications of spectral graph theory to address the inpainting problem of large missing chunks. We used a non-local patch graph representation of the image and proposed a structure detector which leverages the graph rep ...
2014LiveMesh, a tool for real-time rendering of neuronal cells from morphologies
The project goal was to prove the feasibility of GPU-based tessellation to generate neuron membrane mesh representations from parametric descriptions of neurons. The developed prototype software produces a smooth, continuous and high-fidelity representatio ...
2014