{"id":11091,"date":"2023-11-29T13:46:40","date_gmt":"2023-11-29T12:46:40","guid":{"rendered":"https:\/\/www.ibisc.univ-evry.fr\/alexandre-heuillet-will-defend-his-doctoral-thesis-on-monday-december-4-2023-automatic-optimization-of-deep-neural-network-architectures-via-a-differentiable-objective\/"},"modified":"2024-07-06T13:49:32","modified_gmt":"2024-07-06T11:49:32","slug":"alexandre-heuillet-will-defend-his-doctoral-thesis-on-monday-december-4-2023-automatic-optimization-of-deep-neural-network-architectures-via-a-differentiable-objective","status":"publish","type":"post","link":"https:\/\/www.ibisc.univ-evry.fr\/en\/alexandre-heuillet-will-defend-his-doctoral-thesis-on-monday-december-4-2023-automatic-optimization-of-deep-neural-network-architectures-via-a-differentiable-objective\/","title":{"rendered":"Alexandre HEUILLET will defend his doctoral thesis on Monday, December 4, 2023: “Automatic Optimization of Deep Neural Network Architectures via a Differentiable Objective”"},"content":{"rendered":"
\n\t\t\t\t\t
\n\t\t\t\t\t\t

Alexandre HEUILLET defends his doctoral thesis on Monday December 4, 2023 at 3:30 p.m. in Amphitheater Bx30 on the Pelvoux site of the University of \u00c9vry.<\/span><\/p>\n

A broadcast via Zoom is also available:\u00a0https:\/\/universite-paris-saclay-fr.zoom.us\/j\/96288613809?pwd=Uzd5OXJvMHVLdWhWaERtbk9qeXVIdz09<\/strong><\/span><\/a><\/span><\/p>\n

Title<\/strong>\u00a0:\u00a0Exploring Differentiable Neural Architecture Search for Deep Neural Network Design<\/strong><\/h2>\n

Abstract :<\/strong><\/h2>\n

Artificial Intelligence (AI) has gained significant popularity in recent years, primarily due to its successful applications in various domains, including textual data analysis, computer vision, and audio processing. The resurgence of deep learning techniques has played a central role in this success. The groundbreaking paper by Krizhevsky et al., AlexNet, narrowed the gap between human and machine performance in image classification tasks. Subsequent papers such as Xception and ResNet have further solidified deep learning as a leading technique, opening new horizons for the AI community. The success of deep learning lies in its architecture, which is manually designed with expert knowledge and empirical validation. However, these architectures lack the certainty of an optimal solution. To address this issue, recent papers introduced the concept of Neural Architecture Search (NAS), enabling the learning of deep architectures. However, most initial approaches focused on large architectures with specific targets (e.g., supervised learning) and relied on computationally expensive optimization techniques such as reinforcement learning and evolutionary algorithms. In this thesis, we further investigate this idea by exploring automatic deep architecture design, with a particular emphasis on differentiable NAS (DNAS), which represents the current trend in NAS due to its computational efficiency. While our primary focus is on Convolutional Neural Networks (CNNs), we also explore Vision Transformers (ViTs) with the goal of designing cost-effective architectures suitable for real-time applications.<\/p>\n<\/div>

Composition of the doctoral thesis jury<\/span><\/em><\/h3>\n<\/div>\n
\n\n\n\n\n\n\n\n\n\n\n\n\n
Jury member<\/th>\nTitle<\/th>\nAffiliation<\/th>\nFunction in the jury<\/th>\n<\/tr>\n<\/thead>\n
Hichem ARIOUI<\/td>\nAssistant professor<\/td>\nUniversit\u00e9 Paris-Saclay, Univ. \u00c9vry<\/td>\nThesis co-supervisor<\/td>\n<\/tr>\n
Florence D’ALCH\u00c9-BUC<\/td>\nFull professor<\/td>\nT\u00e9l\u00e9com Paris, Institut Polytechnique de Paris<\/td>\nRapporteure<\/td>\n<\/tr>\n
Isabelle GUYON<\/td>\nFull professor<\/td>\nUniversit\u00e9 Paris-Saclay, Google Brain<\/td>\nInvited member<\/td>\n<\/tr>\n
Blaise HANCZAR<\/td>\nFull professor<\/td>\nUniversit\u00e9 Paris-Saclay, Univ. \u00c9vry<\/td>\nExaminateur<\/td>\n<\/tr>\n
David PICARD<\/td>\nFull professor<\/td>\n\u00c9cole des Ponts ParisTech, CNRS, Universit\u00e9 Gustave Eiffel<\/td>\nExaminateur<\/td>\n<\/tr>\n
Ra\u00fal SANTOS-RODRIGUEZ<\/td>\nProfessor<\/td>\nUniversit\u00e9 de Bristol<\/td>\nRapporteur<\/td>\n<\/tr>\n
Hedi TABIA<\/td>\nFull professor<\/td>\nUniversit\u00e9 Paris-Saclay (Univ. \u00c9vry)<\/td>\nThesis supervisor<\/td>\n<\/tr>\n
Kamal YOUCEF-TOUMI<\/td>\nProfessor<\/td>\nMassachusetts Institute of Technology (Etats-Unis)<\/td>\nInvited member<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n<\/div>\n

<\/div>
<\/div>
<\/div>
<\/div>

 <\/p>\n<\/div>

<\/div>
<\/div>
    \n
  • Date: Monday 04\/12\/2023, 3:30pm<\/li>\n
  • Location: Site Pelvoux, University of\u00a0 \u00c9vry, Amphitheater Bx30
    \nWebcast via Zoom https:\/\/universite-paris-saclay-fr.zoom.us\/j\/96288613809?pwd=Uzd5OXJvMHVLdWhWaERtbk9qeXVIdz09<\/li>\n
  • PhD student : Alexandre HEUILLET, IBISC IRA2 team<\/li>\n
  • Thesis supervisors: Hedi TABIA (Full professor Univ. \u00c9vry, IBISC IRA2 team), Hicham ARIOUI (Assistant professor, IBISC SIAM team)<\/li>\n<\/ul>\n<\/div>
    <\/div>\n\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div><\/div><\/div>\n","protected":false},"excerpt":{"rendered":"","protected":false},"author":12,"featured_media":1364,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"inline_featured_image":false,"footnotes":""},"categories":[41,41,161,161,159,162,162,163,163,83,83,52,52,159,169,169],"tags":[],"class_list":{"0":"post-11091","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","6":"hentry","7":"category-in-the-headlines","9":"category-ira2-team","11":"category-siam-team","12":"category-events","14":"category-evra-platform","16":"category-platforms","18":"category-research","21":"category-phd-thesis-defense"},"acf":[],"_links":{"self":[{"href":"https:\/\/www.ibisc.univ-evry.fr\/en\/wp-json\/wp\/v2\/posts\/11091"}],"collection":[{"href":"https:\/\/www.ibisc.univ-evry.fr\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.ibisc.univ-evry.fr\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.ibisc.univ-evry.fr\/en\/wp-json\/wp\/v2\/users\/12"}],"replies":[{"embeddable":true,"href":"https:\/\/www.ibisc.univ-evry.fr\/en\/wp-json\/wp\/v2\/comments?post=11091"}],"version-history":[{"count":7,"href":"https:\/\/www.ibisc.univ-evry.fr\/en\/wp-json\/wp\/v2\/posts\/11091\/revisions"}],"predecessor-version":[{"id":11098,"href":"https:\/\/www.ibisc.univ-evry.fr\/en\/wp-json\/wp\/v2\/posts\/11091\/revisions\/11098"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.ibisc.univ-evry.fr\/en\/wp-json\/wp\/v2\/media\/1364"}],"wp:attachment":[{"href":"https:\/\/www.ibisc.univ-evry.fr\/en\/wp-json\/wp\/v2\/media?parent=11091"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.ibisc.univ-evry.fr\/en\/wp-json\/wp\/v2\/categories?post=11091"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.ibisc.univ-evry.fr\/en\/wp-json\/wp\/v2\/tags?post=11091"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}