Leveraging Domain Adaptation in Unsupervised Federated Learning

il y a 2 jours


Rouen, France University of Rouen Normandy Temps plein

Topic description Context Federated learning (FL) enables models to learn from distributed datasets across diverse clients (e.g., edge devices, hospitals, or industrial sites) while maintaining privacy [1]. A major challenge in supervised FL is statistical variability, which causes client distribution shifts—differences in data distributions across clients [2]. Additionally, many clients contain unlabeled data that remain unused due to annotation costs, inconsistent labeling protocols, and model updates. As a result, FL with unlabeled data has gained attention, with approaches that take advantage of clustering techniques to correct client shifts [3,4]. Since domain shifts in Domain Adaptation (DA) are analogous to client shifts in federated learning, federated domain adaptation has naturally emerged to transfer methodological advances from DA to address distribution shifts [5]. DA seeks to learn a domain-invariant feature space between labeled source and target domains, whereas unsupervised domain adaptation handles the more challenging case where target data are unlabeled [6]. Recent studies have incorporated DA techniques into FL, primarily in centralized settings (e.g., at the server) to address evolving distribution shifts [7]. Among these methods, Source-Free Domain Adaptation (SFDA) presents a promising direction, as it enables a client to share a trained model with another party, allowing the recipient to perform unsupervised adaptation using their own data in conjunction with the shared model [8]. Finally, performance degradation and poor generalization are further exacerbated by class imbalance in unlabeled data, where majority classes dominate while minority classes are underrepresented. In supervised FL, this issue can be mitigated by estimating clients’ class distributions and applying loss-based adjustments [9]. However, such strategies are not directly applicable in unsupervised settings, where class labels are unavailable. Objectives The main objectives of this internship are as follows: first, to get familiar with an already developed federated learning framework that includes the privacy-preserving Source-Free Domain Adaptation (SFDA) approach. Second, to study the effect of class imbalance on performance in federated learning. Third, to investigate methods used in supervised domain adaptation and federated learning to address class imbalance [1,2]. Finally, based on the insights from the previous objectives, the candidate will propose new strategies. The experiments will be conducted on publicly available time-series datasets, including sensor data from accelerometers, gyroscopes, and body sensors collected from multiple subjects (e.g., Human Activity Recognition and EEG signal datasets). Additionally, experiments may be performed on well-known image classification datasets to further evaluate the proposed approach. Starting date -09-01 Funding category Public/private mixed funding Funding further details



  • Rouen, France University of Rouen Normandy Temps plein

    **Leveraging Domain Adaptation in Unsupervised Federated Learning**: - Réf **ABG-134073** - Stage master 2 / Ingénieur - Durée 6 mois - Salaire net mensuel 630 - 28/10/2025 - University of Rouen Normandy - Lieu de travail- Rouen Normandie France - Champs scientifiques- Science de la donnée (stockage, sécurité, mesure, analyse) - Mathématiques - Mots...


  • Rouen, Normandie, France Walter Learning Temps plein

    L'école Walter Learning conçoit, produit et dispense des formations en ligne à destination des professionnels et des personnes qui cherchent à construire leur avenir professionnel.Alliant création de contenus pédagogiques et expertise technologique, nous produisons nos formations en interne avec des spécialistes reconnus, et les diffusons sur notre...


  • Rouen, France CHU Rouen Temps plein

    Participation à la rédaction et/ou à la mise à jour de process dans le domaine de la Formation en lien direct avec l’encadrement du pôle Formation. - Appui au projet de dématérialisation des demandes d’inscription à une formation des agents du CHU - Appui à la gestion des dossiers relevant des dispositifs individuels de formation (Congé de...

  • Neuropsychologue H/F

    il y a 1 semaine


    Rouen, Normandie, France Select-in Temps plein

    Présentation de la société Présentation : Cabinet de recrutement et d'approche directe au rayonnement national, SELECT-IN accompagne associations, fondations et collectivités sur leurs recrutements de profils cadres et experts métiers.SELECT-IN accompagne son client, acteur reconnu du secteur médico-social, dans le recrutement d'un Neuropsychologue...

  • Neuropsychologue H/F

    il y a 2 semaines


    Rouen, France SELECT-IN Temps plein

    Neuropsychologue H/F SELECT-IN•Rouen, Normandie, France Description de poste INFOS CLÉS : NEUROPSYCHOLOGUE - ÉTABLISSEMENT MÉDICO-SOCIAL ADULTES - CDI TEMPS PARTIEL (0,7 ETP) - ROUEN (76) Le poste est basé dans un établissement médico-social à proximité de Rouen accueillant des adultes en situation de polyhandicap. Vos principales missions sont les...

  • Neuropsychologue H/F

    il y a 2 semaines


    Rouen, France SELECT-IN Temps plein

    INFOS CLÉS : NEUROPSYCHOLOGUE - ÉTABLISSEMENT MÉDICO-SOCIAL ADULTES - CDI TEMPS PARTIEL (0,7 ETP) - ROUEN (76) Le poste est basé dans un établissement médico-social à proximité de Rouen accueillant des adultes en situation de polyhandicap. Vos principales missions sont les suivantes : - Réaliser des évaluations cognitives et neuropsychologiques. -...


  • Rouen, France in&ma Temps plein

    Vous êtes **actuellement en Bac +2/+3** et souhaitez poursuivre vos études afin d’évoluer vers un poste à responsabilités dans les domaines de **:la gestion de projets, la maintenance, la production, la qualité, les méthodes, la supply chain, le lean management, l’ingénierie** In&ma, Ecole Supérieure de la Performance, propose depuis 30 ans,...

  • Software Engineer Domain

    il y a 5 jours


    Paris / Lille / Bordeaux / Lyon / Rennes / Rouen / Toulouse, France Scaleway Temps plein

    OUR STORY: Join Scaleway and shape the sovereign cloud of tomorrow Since 1999, we have been designing secure, sustainable infrastructures aimed at supporting the most ambitious companies. Historically known for our dedicated servers (Dedibox), we made a strategic shift to cloud computing in 2015. Staying true to our principles of simplicity, flexibility,...

  • HR Business Partner

    il y a 2 semaines


    Rouen, France Allianz Direct Versicherungs-AG French b Temps plein

    HR Business Partner Location: Mont Saint Aignan or Paris (with regular travel between these three sites) As a Human Resources Business Partner (HRBP), you will play a crucial role in supporting Allianz Direct's strategic growth and transformation. With a focus on managerial support, key HR processes, and integration specifics, you will lead major strategic...

  • HR Business Partner

    il y a 2 semaines


    Rouen, France Allianz Temps plein

    Retour aux résultats Société Allianz Lieu(x) Rouen, Paris Cadre Ressources humaines / Formation Publiée depuis 1 jour Die Allianz Gruppe gehört zu den größten und renommiertesten Finanzdienstleistern der Welt und bietet ein internationales Netzwerk bekannter Marken mit erstklassigen Produkten. Vor allem streben wir nach Spitzenleistungen in unserem...