Matching Convolutional Neural Networks without Priors about Data

Lassance, Carlos; Vialatte, J; Gripon, Vincent

doi:10.1109/dsw.2018.8439900

Cited by 2 publications

(10 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, it is tough to do so in practice [91]. We will explore this drawback in more detail in Chapter 4.…”

Section: Fully Connected Layer (Fc)mentioning

confidence: 99%

“…In [91], we introduce the use of the Pines dataset in the context of deep neural networks for the first time. The task of the PINES dataset is to identify the emotional rating a subject gives to a picture by using the fMRI scan of its brain.…”

Section: Pinesmentioning

confidence: 99%

“…In order to transfer this same translation concept from euclidean spaces to graphs, we choose to use in this work the translation definition and inference methods introduced in [118] that we extend in [91].…”

Section: Translations On Graphsmentioning

confidence: 99%

See 2 more Smart Citations

Graphs for deep learning representations

Lassance¹

2020

Preprint

View full text Add to dashboard Cite

Ces dernières années, les méthodes d'apprentissage profond ont atteint l'état de l'art dans une vaste gamme de tâches d'apprentissage automatique, y compris la classification d'images et la traduction automatique. Ces architectures sont assemblées pour résoudre des tâches d'apprentissage automatique de bout en bout. Afin d'atteindre des performances de haut niveau, ces architectures nécessitent souvent d'un très grand nombre de paramètres.Les conséquences indésirables sont multiples, et pour y remédier, il est souhaitable de pouvoir compreendre ce qui se passe à l'intérieur des architectures d'apprentissage profond.Il est difficile de le faire en raison de: i) la dimension élevée des représentations, and ii) la stochasticité du processus de formation. Dans cette thèse, nous étudions ces architectures en introduisant un formalisme à base de graphes, s'appuyant notamment sur les récents progrès du traitement de signaux sur graphe (TSG). À savoir, nous utilisons des graphes pour représenter les espaces latents des réseaux neuronaux profonds. Nous montrons que ce formalisme des graphes nous permet de répondre à diverses questions, notamment: i) mesurer des capacités de généralisation, ii) réduire la quantité de des choix arbitraires dans la conception du processus d'apprentissage, iii) améliorer la robustesse aux petites ix x RÉSUMÉ perturbations ajoutées sur les entrées, and iv) réduire la complexité des calculs.

show abstract

“…However, it is tough to do so in practice [91]. We will explore this drawback in more detail in Chapter 4.…”

Section: Fully Connected Layer (Fc)mentioning

confidence: 99%

Section: Pinesmentioning

confidence: 99%

See 1 more Smart Citation

Graphs for deep learning representations

Lassance¹

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Data is back projected using only those first eigenvectors, and a low-dimensional representation of the signal is obtained. However, the analogy with the FT is not complete, as for instance, the GFT using a 2-dimensional grid does not match the 2D classical FT. 7 Finding the best graph structure for a given problem is a question that recently sparked a lot of interest in the literature. [14][15][16] In many cases, a good solution consists in using the covariance matrix (or its inverse) as the adjacency matrix of the graph.…”

Section: Related Workmentioning

confidence: 99%

“…6 However, it is known that GFT can fail at fully leveraging the underlying structure, even in the case of very regular domains. 7 The question we ask in this paper is: what are the linear transforms that are the best fitted to predict the evolution of GTS?…”

Section: Introductionmentioning

confidence: 99%

Comparing linear structure-based and data-driven latent spatial representations for sequence prediction

Bontonou

Lassance

Gripon

et al. 2019

Wavelets and Sparsity XVIII

Self Cite

View full text Add to dashboard Cite

Predicting the future of Graph-supported Time Series (GTS) is a key challenge in many domains, such as climate monitoring, finance or neuroimaging. Yet it is a highly difficult problem as it requires to account jointly for time and graph (spatial) dependencies. To simplify this process, it is common to use a two-step procedure in which spatial and time dependencies are dealt with separately. In this paper, we are interested in comparing various linear spatial representations, namely structure-based ones and data-driven ones, in terms of how they help predict the future of GTS. To that end, we perform experiments with various datasets including spontaneous brain activity and raw videos.

show abstract

Matching Convolutional Neural Networks without Priors about Data

Cited by 2 publications

References 12 publications

Graphs for deep learning representations

Graphs for deep learning representations

Comparing linear structure-based and data-driven latent spatial representations for sequence prediction

Contact Info

Product

Resources

About