[{"command":"insert","method":null,"selector":".course-modal-ajax-placeholder-964172","data":"\u003Cdiv class=\u0022container content-rte\u0022\u003E\n \u003Cdiv class=\u0022row\u0022\u003E\n \u003Cdiv class=\u0022box col-md-6 bg-primary text-white\u0022\u003E\n \u003Ch2 class=\u0022h1 text-left\u0022\u003E Bases, concepts et enjeux du Deep Learning\n\u003C\/h2\u003E\n \u003Cdiv class=\u0022field-spacing\u0022\u003E\n \u003Cdiv class=\u0022field__label h3 dash\u0022\u003EECTS :\n 2.5\n \u003C\/div\u003E\n \u003C\/div\u003E\n \u003Cdiv class=\u0022field-spacing\u0022\u003E\n \u003Cdiv class=\u0022field__label h3 dash\u0022\u003ESemestre calendaire :\n Semestre 1\n \u003C\/div\u003E\n \u003C\/div\u003E\n \u003Cdiv class=\u0022field-spacing\u0022\u003E\n \u003Cdiv class=\u0022field__label h3 dash\u0022\u003ED\u00e9tail du volume horaire\u003C\/div\u003E\n \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n \u003Cspan class=\u0022field__label h4 mt-5 \u0022\u003ECours magistraux :\u003C\/span\u003E\n \u003Cspan\u003E4\u003C\/span\u003E\n \u003C\/div\u003E\n \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n \u003Cspan class=\u0022field__label h4 mt-5 \u0022\u003ETravaux pratiques :\u003C\/span\u003E\n \u003Cspan\u003E14\u003C\/span\u003E\n \u003C\/div\u003E\n \u003C\/div\u003E\n \u003Cdiv class=\u0022field-spacing\u0022\u003E\n \u003Cdiv class=\u0022field__label h3 dash\u0022\u003ELangue d\u0026#039;enseignement\u003C\/div\u003E\n \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n \u003Cspan\u003EFran\u00e7ais\u003C\/span\u003E\n \u003C\/div\u003E\n \u003C\/div\u003E\n \u003Cdiv class=\u0022field-spacing\u0022\u003E\n \u003Cdiv class=\u0022field__label h3 dash\u0022\u003EEnseignement \u00e0 distance\u003C\/div\u003E\n \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n \u003Cspan\u003Enon\u003C\/span\u003E\n \u003C\/div\u003E\n \u003C\/div\u003E\n \u003C\/div\u003E\n \u003Cdiv class=\u0022box col-md-6\u0022\u003E\n \u003Cdiv class=\u0022field-spacing\u0022\u003E\n \u003Cdiv class=\u0022field__label h3 dash\u0022\u003EPr\u00e9requis\u003C\/div\u003E\n \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n \u003Cspan\u003E\u003Cp\u003Enotions de base en alg\u00e8bre lin\u00e9aire, en physique et en chimie des mat\u00e9riaux\u0026nbsp;; langage python\u003C\/p\u003E\u003C\/span\u003E\n \u003C\/div\u003E\n \u003C\/div\u003E\n \u003Cdiv class=\u0022field-spacing\u0022\u003E\n \u003Cdiv class=\u0022field__label h3 dash\u0022\u003EProgramme \/ plan \/ contenus\u003C\/div\u003E\n \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n \u003Cspan\u003E\u003Cp\u003ELes \u00e9tudiants explorent les grandes familles de mod\u00e8les\u0026nbsp;: des r\u00e9seaux convolutifs (CNN) utilis\u00e9s pour l\u2019analyse d\u2019images aux r\u00e9seaux de graphes (GNN) capables de repr\u00e9senter des structures atomiques, jusqu\u2019aux transformeurs et aux grands mod\u00e8les de langage (LLM) qui ouvrent la voie \u00e0 l\u2019IA g\u00e9n\u00e9rative. L\u2019accent est mis sur la compr\u00e9hension intuitive des concepts : comment un r\u00e9seau apprend, comment il s\u2019adapte et comment il peut r\u00e9v\u00e9ler des r\u00e9gularit\u00e9s cach\u00e9es dans les donn\u00e9es scientifiques. Le but est de conna\u00eetre les champs d\u2019applications de chaque m\u00e9thode.\u003C\/p\u003E\n\u003Cp\u003ELes TP et le mini-projet abordent la lecture\/visualisation de donn\u00e9es (CSV, HDF5) ; les r\u00e9seaux de neurones (MLP) : construction d\u2019un mod\u00e8le simple (PyTorch\/TensorFlow), fonctions de co\u00fbt et optimisation, \u00e9tude d\u2019hyperparam\u00e8tres, validation crois\u00e9e et m\u00e9triques, application \u00e0 la r\u00e9gression de propri\u00e9t\u00e9s de mat\u00e9riaux\u0026nbsp;; les repr\u00e9sentations structur\u00e9es : introduction aux CNN (images) et GNN (graphes atomiques, pytorch-geometric), notions d\u2019invariance et de repr\u00e9sentation physique\u0026nbsp;; les mod\u00e8les g\u00e9n\u00e9ratifs.\u003C\/p\u003E\u003C\/span\u003E\n \u003C\/div\u003E\n \u003C\/div\u003E\n \u003Cdiv class=\u0022field-spacing\u0022\u003E\n \u003Cdiv class=\u0022field__label h3 dash\u0022\u003EObjectifs d\u0026#039;apprentissage\u003C\/div\u003E\n \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n \u003Cspan\u003E\u003Cp\u003ECette UE invite \u00e0 un voyage au c\u0153ur de l\u2019intelligence artificielle moderne, le deep learning. On y d\u00e9couvre comment les m\u00e9thodes classiques d\u2019apprentissage automatique ont \u00e9volu\u00e9 vers les r\u00e9seaux de neurones profonds, qui transforment aujourd\u2019hui la recherche scientifique.\u003C\/p\u003E\u003C\/span\u003E\n \u003C\/div\u003E\n \u003C\/div\u003E\n \u003Cdiv class=\u0022field-spacing\u0022\u003E\n \u003Cdiv class=\u0022field__label h3 dash\u0022\u003EBibliographie\u003C\/div\u003E\n \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n \u003Cspan\u003E\u003Cul\u003E\n\u003Cli\u003EIan Goodfellow, Yoshoua Bengio, Aaron Courville, Deep Learning, MIT Press, 2016.\u003C\/li\u003E\n\u003Cli\u003EAur\u00e9lien G\u00e9ron, Hands-On Machine Learning with Scikit-Learn, Keras \u0026amp; TensorFlow, O\u2019Reilly, 2022\u003C\/li\u003E\n\u003C\/ul\u003E\u003C\/span\u003E\n \u003C\/div\u003E\n \u003C\/div\u003E\n \u003Cdiv class=\u0022field-spacing\u0022\u003E\n \u003Cdiv class=\u0022field__label h3 dash\u0022\u003ENature de l\u0026#039;\u00e9valuation\u003C\/div\u003E\n \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n \u003Cspan\u003EEvaluation Continue non Int\u00e9grale\u003C\/span\u003E\n \u003C\/div\u003E\n \u003C\/div\u003E\n \u003C\/div\u003E\n \u003C\/div\u003E\n\u003C\/div\u003E\n","settings":null}]