Výsledky vyhledávání - "Vincent Roulet"

Discriminative clustering with representation learning with any ratio of labeled to unlabeled data

Autor: Corinne Jones, Vincent Roulet, Zaid Harchaoui

Publikováno v: Statistics and Computing. 32

We present a discriminative clustering approach in which the feature representation can be learned from data and moreover leverage labeled data. Representation learning can give a similarity-based clustering method the ability to automatically adapt

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::2d9c4caed7f7c7d196e195e4c870811c
https://doi.org/10.1007/s11222-021-10067-x

Zobrazit plný text záznamu

On the Smoothing of Deep Networks

Autor: Vincent Roulet, Zaid Harchaoui

Publikováno v: CISS

Many popular deep neural networks implement an input-output mapping that is non-smooth with respect to the network parameters. This non-smoothness may have contributed to the difficulty of analyzing deep learning theoretically. Sophisticated approach

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::2430e7c27330b0a8e22695b987bfe1c9
https://doi.org/10.1109/ciss50987.2021.9400285

Zobrazit plný text záznamu

Sharpness, Restart and Acceleration

Autor: Alexandre d'Aspremont, Vincent Roulet

Publikováno v: SIAM Journal on Optimization
SIAM Journal on Optimization, Society for Industrial and Applied Mathematics, 2020, 30 (1), pp.262-289. ⟨10.1137/18M1224568⟩
SIAM Journal on Optimization, 2020, 30 (1), pp.262-289. ⟨10.1137/18M1224568⟩

The {\L}ojasiewicz inequality shows that sharpness bounds on the minimum of convex optimization problems hold almost generically. Sharpness directly controls the performance of restart schemes, as observed by Nemirovsky and Nesterov (1985). The const

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::db2ddee5ddd406e4f4694edb08385007
https://hal.archives-ouvertes.fr/hal-02983236

Zobrazit plný text záznamu

Differentiable Programming à la Moreau

Autor: Vincent Roulet, Zaid Harchaoui

The notion of a Moreau envelope is central to the analysis of first-order optimization algorithms for machine learning. Yet, it has not been developed and extended to be applied to a deep network and, more broadly, to a machine learning system with a

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::62682b43ed582035e98dcad93d1e8663

Zobrazit plný text záznamu

An Elementary Approach to Convergence Guarantees of Optimization Algorithms for Deep Networks

Autor: Vincent Roulet, Zaid Harchaoui

Publikováno v: Allerton

We present an approach to obtain convergence guarantees of optimization algorithms for deep networks based on elementary arguments and computations. The convergence analysis revolves around the analytical and computational structures of optimization

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::30654d15a5b5ec37660e04af045440e1

Zobrazit plný text záznamu

On the Convergence of the Iterative Linear Exponential Quadratic Gaussian Algorithm to Stationary Points

Autor: Zaid Harchaoui, Vincent Roulet, Siddhartha S. Srinivasa, Maryam Fazel

Publikováno v: ACC

A classical method for risk-sensitive nonlinear control is the iterative linear exponential quadratic Gaussian algorithm. We present its convergence analysis from a first-order optimization viewpoint. We identify the objective that the algorithm actu

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::95b26afe7fd8360dae108db8821b38dd
http://arxiv.org/abs/1910.08221

Zobrazit plný text záznamu

Aktualne zagadnienia odpowiedzialności w prawie pracy

Autor: Vincent Roulet

Gra w odpowiedzialność nie jest całkowicie jednokierunkowa. Pracownicy, związki zawodowe i państwo również ponoszą odpowiedzialność na mocy prawa pracy; czasami akcjonariusze - jeśli nie pracodawcy - wykorzystywali subtelności prawa, aby

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::23d0d3ee22e8e6b9836bee7839cafdce
https://hdl.handle.net/11089/42047

Zobrazit plný text záznamu

An efficient computational strategy for composite laminates assemblies including variability

Autor: Vincent Roulet, Laurent Champaney, Pierre-Alain Boucard

Publikováno v: International Journal of Solids and Structures
International Journal of Solids and Structures, Elsevier, 2013, 50 (18), pp.2749-2757. ⟨10.1016/j.ijsolstr.2013.04.028⟩

The aim of this work is to present an efficient numerical strategy for studying the influence of the material parameters on problems involving 3D assemblies of composite parts with contact and friction. This approach is based on the multiscale LATIN

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::a82e4e5e63d08a34e907de9136b9ef81

Zobrazit plný text záznamu

Quelques propos introductifs à la réforme du droit des contrats, du régime général et de la preuve des obligations

Autor: Vincent Roulet

Publikováno v: Périodiques Scientifiques en Édition Électronique.

Roulet Vincent. Quelques propos introductifs à la réforme du droit des contrats, du régime général et de la preuve des obligations. In: Revue juridique de l'Ouest, 2016-4. pp. 57-66.

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::81fd7f7060687fc51a5a7fef8521bc9e
https://www.persee.fr/doc/juro_0990-1027_2016_num_29_4_4886

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání