Institut de Mathématiques de Marseille, UMR 7373




Rechercher


Accueil >

11 janvier 2019: 2 événements

Séminaire

  • Séminaire Teich

    Vendredi 11 janvier 11:00-13:30 - Colin GUILLARMOU - LMO, Orsay

    Séminaire Teich (TBA)

    Résumé : TBA
    Exposé commun avec le séminaire de probabilités

    JPEG - 4.5 ko
    Colin GUILLARMOU

    Lieu : FRUMAM - Aix-Marseille Université - Site St Charles
    3, place Victor Hugo - case 39
    13331 MARSEILLE Cedex 03

    Exporter cet événement
    Document(s) associé(s) :

    En savoir plus : Séminaire Teich

  • Séminaire Signal et Apprentissage

    Vendredi 11 janvier 14:00-15:00 - Nick KINGSBURY - University of Cambridge

    Combining Complex Wavelets with Deep Networks : aiming to improve learning efficiency for vision systems

    Résumé : Scattering networks [Bruna & Mallat, IEEE Trans PAMI 2013 ; Oyallon & Mallat, CVPR 2015] may be interpreted as convolutional network layers in which the filters are defined by complex wavelet transforms and whose layer non-linearities are typically complex modulus (L2-norm) operators. Usually they are pre-designed using standard complex wavelet design methodologies that are based on accumulated human knowledge about vision systems, and they involve minimal training. It is found that several layers of scatternet can usefully replace the early layers of a deep convolution neural net (CNN). The aim of this strategy is that the deterministic and complete nature of the wavelet transformations will result in deep networks that are faster at learning, more comprehensible in their behaviour and perhaps better at generalisation than a CNN which has to learn all of its layers from finite amounts of training data. Furthermore, by employing tight-frame overcomplete wavelets and L2-norm nonlinearities, signal energy may be conserved through the scatternet layers, leading to some interesting strategies for subspace selection.
    -
    In this talk we shall suggest a number of ways that dual-tree complex wavelets may be incorporated into deep networks, either to generate scatternet front-ends or to produce interesting alternatives to standard convolutional layers, embedded deeper in the network. We will also show how recent ideas on CNN layer visualisation can be extended to include the wavelet-based layers too. We shall pose more questions than answers, while also presenting a few results from current stages of this work. I am very grateful to my co-researchers on this project, Amarjot Singh and Fergal Cotter.



    JPEG - 11 ko
    Nick KINGSBURY

    Lieu : CMI, salle de séminaire R164 (1er étage) - I2M - Château-Gombert
    39 rue Frédéric Joliot-Curie
    13453 MARSEILLE cedex 13

    Exporter cet événement
    Document(s) associé(s) :

    En savoir plus : Séminaire Signal et Apprentissage

11 janvier 2019: 1 événement

groupe de travail

11 janvier 2019: 1 événement

Manifestation scientifique