À propos de Gautier
Gautier Cosne - Data Scientist Expert en Signal Processing et Machine Learning
Data Scientist passionné par et spécialisé dans le traitement du signal et l'apprentissage automatique, avec une expertise particulière dans le domaine de la santé numérique et des technologies portables. Fort d'une formation pointue en mathématiques appliquées et en informatique de l'IMT Atlantique, je combine rigueur académique et expérience pratique pour relever des défis complexes en science des données.
Expertise clé
- Science des données et Machine Learning: Conception et implémentation d'algorithmes avancés pour l'analyse de données complexes et volumineuses.
- Traitement du signal: Expert en extraction de mesures pertinentes à partir de signaux physiologiques et de capteurs portables (IMU, accéléromètres).
- Santé numérique: Expérience approfondie dans l'application de la science des données aux maladies neurologiques et à la santé cognitive.
- Deep Learning et IA générative: Compétences avancées en deep learning, incluant le travail sur des modèles génératifs de pointe.
Proposition de valeur
- Une expertise de pointe en science des données et traitement du signal, particulièrement adaptée aux projets de santé numérique.
- Une capacité éprouvée à traduire des concepts techniques complexes en insights actionnables pour des équipes pluridisciplinaires.
- Une approche innovante combinant les dernières avancées en IA et en apprentissage automatique avec une solide compréhension des défis du monde réel.
- Un engagement à livrer des solutions robustes, évolutives et alignées sur les objectifs stratégiques de votre entreprise.
Prêt à relever vos défis les plus complexes en matière de données et à propulser vos projets vers de nouveaux sommets d'innovation et d'efficacité.
Français
Bilingue ou natif
Anglais
Bilingue ou natif
Accepte de travailler sur site
Paris (jusqu’à 40 km)
Expériences
- BIOGEN MA INCData ScientistINDUSTRIE PHARMACEUTIQUEjuin 2020 - Aujourd'hui (6 ans)Paris, FranceData Scientist | Neuroscience | Biogen Digital HealthSeasoned Data Scientist with a proven track record of driving innovation at the intersection of data science, signal processing, and healthcare. My expertise lies in developing digital measures from physiological signals, deploying advanced machine learning techniques, and delivering impactful solutions that align with business objectives.At Biogen Digital Health, I have successfully accelerated key studies by leveraging my technical skills and leadership abilities to orchestrate cross-functional collaborations and ensure seamless project execution. My ability to bridge technical knowledge and business acumen has consistently resulted in actionable insights and publication-ready analyses.Passionate about
- Developing digital measures from physiological signals
- Deploying advanced machine learning techniques
- Pushing the boundaries of healthcare technology
- Driving real-world impact through data-driven solutions
Published work- About DISPEL, an open-source Python library that standardizes the extraction of sensor-derived measures from IMUs. https://ieeexplore.ieee.org/abstract/document/10533679 DOI: 10.1109/OJEMB.2024.3402531 https://github.com/newcastleuniversity/DISPEL
- About extracting relevant sensor-derived measures from wearable devices, particularly IMUs for adults with neurological disorder. https://content.iospress.com/articles/journal-of-neuromuscular-diseases/jnd240004 DOI: 10.3233/JND-240004
- About analyzing large-scale data from iPhones and Apple Watches to classify mild cognitive impairment. https://www.researchsquare.com/article/rs-4173311/v1 DOI:10.21203/rs.3.rs-4173311/v1
Keywords: Data Science, Neuroscience, Healthcare, Machine Learning, Signal Processing, Digital Health, Biogen, Data-Driven Solutions, Innovation. - Pierre Orban - Research Center of the University Institute in Mental Health of Montreal
Sur Malt
Machine Learning EngineerSECTEUR MÉDICALavril 2020 - juin 2020 (3 mois)Montréal, CanadaMachine Learning EngineerCollaborated with Dr. Pierre Orban, an expert in functional magnetic resonance imaging (fMRI) and schizophrenia research, and Prof. Alejandro Murua, a specialist in machine learning and statistical methods, to develop a novel approach for identifying schizophrenia subtypes through brain connectivity data.Custom Machine Learning Model Development- Designed and implemented a bespoke Gaussian Mixture Model (GMM) with additive effects, tailored specifically for schizophrenia subtype identification.
- Innovated on traditional GMMs by incorporating an additive component to model the superposition of healthy brain functions and disease states.
Technical Skills Demonstrated- Advanced Python programming, with a focus on scientific computing libraries (NumPy, SciPy, Scikit-learn)
- Custom implementation of machine learning algorithms
- Statistical modeling and Bayesian inference
- Version control and collaborative coding practices
- MILADeep Learning | Visiting ResearcherCENTRES DE RECHERCHEmai 2019 - février 2020 (9 mois)Montreal, CanadaDeep Learning Research | Mila, Montreal, CanadaDuring my year at Mila, I had the opportunity to collaborate with a world-class team of deep learning researchers under the guidance of Turing Laureate Yoshua Bengio. My primary focus was on the "This Climate Does Not Exist" project, an innovative generative AI-driven tool designed to raise awareness about the impacts of climate change.Technical ContributionsAs part of this interdisciplinary team, I played a key role in implementing state-of-the-art Generative Adversarial Networks (GANs) to develop a Flooding Simulator. This involved extending the GAN architecture to generate realistic visualizations of climate-related disasters, such as floods and wildfires, by analyzing and transforming Google Street View images. Through this work, I gained extensive experience in deep learning, computer vision, and neural network architecture.Impact and RecognitionOur project successfully bridged the gap between AI and climate science, harnessing machine learning to incite action. The tool received recognition at the NeurIPS Workshop and Montreal AI Symposium, showcasing its potential to use technology for social good.LearningsThis experience honed my skills in advanced machine learning techniques, collaboration in a multidisciplinary environment, and effectively communicating complex technical concepts.Published work:
- Developed ClimateGAN, a deep learning model for visualizing extreme climate events, which was presented at ICLR 2022. ClimateGAN, is a model that leverages both simulated and real data for unsupervised domain adaptation and conditional image generation. https://openreview.net/forum?id=EZNOb_uNpJk - DOI: 10.48550/arXiv.2110.02871 Simulator available online: https://thisclimatedoesnotexist.com/
- Worked on establishing evaluation metrics for climate change image realism: https://iopscience.iop.org/article/10.1088/2632-2153/ab7657 - DOI: 10.1088/2632-2153/ab7657
Avis
Recommandations
Soyez le premier à recommander Gautier
Contribuez à la réussite de ce freelance en partageant votre expérience de collaboration avec lui.
Ces profils de freelance correspondent également à vos critères
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Formations
- Master's degree, Information Processing and Machine LearningIMT Atlantique2019Machine Learning, Deep Learning, Computer Vision Remote Sensing, Information processing Computer Science, Software and Data Engineering Operation Research, Mathematics and Signal Processing Bases in Computer Networks
Certifications
- Databricks Certified Machine Learning AssociateDatabricks2023
- Neural Networks and Deep LearningCoursera2019