Offre en lien avec l’Action/le Réseau : – — –/– — –

Laboratoire/Entreprise : Orange Lab / UCA-Inria-CNRS MAASAI Team
Durée : 36 months
Contact : michel.riveill@univ-cotedazur.fr
Date limite de publication : 2022-03-14

Contexte :
TThe goal of the thesis is to perform research on decentralized and efficient federated AutoML learning for heterogeneous embedded devices.

The training of AI models for service delivery is today facing a conceptual transformation, by shifting the learning of models close to the data, embedded on users’ devices. These devices have limited resources and must remain fully operational during the learning phase. In addition, users today generate sensitive data and new collaborative algorithms for learning need to be developed and optimized for different embedded devices, ranging from smartphones to IoT.

Nowadays, to build an AI model it is necessary to collect data on a central server (cloud). The problems of this method are related to privacy, control of data usage and computational resources. Federated learning (FL) [1,2] is a new AI approach with collaborative training that resolves these problems. Models are trained on local users’ data and its parameters only are exchanged with other users to build a global model. The challenges of Federated Learning are (a) obtaining efficient and robust decentralized FL models with heterogeneous data (b) optimizing resources for actual operational deployment and (c) customizing services and optimizing model based on available resources for groups of users, because a single global model may be less explainable, accurate and appropriate when compared to a personalized model.

We will deploy deep neural networks on users’ devices because they have high classification/prediction accuracy in various tasks. However, their training requires a significant effort in terms of finding optimal hyperparameters, which limits their use at devices with constrained resources. Emerging areas address the problem of automatic neural network generation [3] and automatic search for appropriate architectures (Neural Architecture Search-NAS), features required for real-world deployments. FL NAS [4] aims at optimizing the architecture of neural network models in the FL environment. Many questions in this domain remain open. For example, there are no approaches developed for FL with clients having the same sample space and a different feature space.

Sujet :
The objective of the thesis is to (a) design a federated learning framework to automatically generate low-power neural networks in compliance with GDPR [5] with homogeneous (b) and heterogeneous devices under device constraints (availability, resources, states) and to study it in a fully decentralized Peer-to-Peer federated learning setup.

[1] J. Konecny, H. B. McMahan, F. X. Yu, P. Richtarik, A. T. Suresh and D. Bacon, “Federated Learning: Strategies for Improving Communication Efficiency,” in arXiv, 2017, pp. 1-10.

[2] K. Bonawitz, H. Eichner, W. Grieskamp, D. Huba, A. Ingerman, V. Ivanov, Ch. M. Kiddon, J. Konečný, S. Mazzocchi, B. McMahan, T. Van Overveldt, D. Petrou, D. Ramage and J. Roselander, “Towards Federated Learning at Scale: System Design,” SysML 2019, https://arxiv.org/abs/1902.01046, 2019.

[3] A. Wong, M. J. Shafiee, B. Chwyl and F. Li, “FermiNets: Learning generative machines to generate efficient neural networks via generative synthesis,” 1809.05989.pdf (arxiv.org), NIPS, 2018.

[4] H. Zhu, H. Zhang and Y. Jin, “From Federated Learning to Federated Neural Architecture Search,” https://arxiv.org/pdf/2009.05868.pdf, 2020.

[5] Regulation (EU) 2016/679 of the European Parliament and of the Council (article 30), https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:32016R0679&from=EN#d1e3265-1-1, Archived from the original on 28 June 2017.

Profil du candidat :
You have a Masters degree in Data Science or Computer Science and you are a curious person that likes to learn and seek for solutions. You are highly motivated to do your thesis in the emerging field of distributed algorithms for embedded devices. You have skills in machine learning, optimization and statistics (essential) as well as good programming skills and knowledge in the field of embedded devices (desirable). Interest in the field of Signal Processing is a plus.

Furthermore, autonomy and open-mindedness are the qualities particularly appreciated for research work. The dynamism, the strength of proposal and the capacities of communication are also required for this position. English will be used throughout the thesis (reading state of the art, writing articles and presenting results at international conferences) and excellent level of English is therefore required.

Formation et compétences requises :
Master or Ecole d’ingénieur.

Adresse d’emploi :
Orange Lab Sophia Antipolis.

Contact :
– Tamara.TOSIC@orange.com,
– Michel.RIVEILL@univ-cotedazur.fr

Offre en lien avec l’Action/le Réseau : Musiscale/– — –

Laboratoire/Entreprise : CRIStAL, Inria, Université de Lille
Durée : 3 ans
Contact : louis.bigo@univ-lille.fr
Date limite de publication : 2022-03-14

Contexte :
Depuis une dizaine d’années, les réseaux de neurones profonds font l’objet de nombreuses recherches dans le domaine du traitement automatique du langage naturel (Natural Language Processing). Ces recherches ont de multiples applications allant de l’analyse de corpus à la génération automatique de contenu.

La nature temporelle de la musique encourage et facilite sa représentation sous la forme de séquences d’éléments à différentes échelles, généralement des accords ou des notes, comparables à des séquences de mots. Cette séquentialité, ainsi que l’assimilation courante de la musique à une sorte de langage, ont motivé l’utilisation d’outils originalement conçus pour des tâches de NLP pour le traitement automatique de données musicales (Music Information Retrieval) pour des tâches variées incluant l’analyse et la génération automatique de musique.

Sujet :
L’objectif central de cette thèse est d’évaluer l’adaptabilité, la performance et la pertinence de techniques de NLP lorsqu’elles sont appliquées sur des données musicales. On se concentrera en particulier sur l’application en musique de trois principes essentiels du NLP :

* le principe d’attention mutuelle (self-attention)
* la segmentation (tokenization)
* l’apprentissage par transfert (transfert learning)

Ces principes seront étudiés à travers l’entraînement et l’évaluation de modèles musicaux inspirés par des modèles majeurs de NLP incluant l’auto-encodeur BERT ou le modèle auto-regressif GPT. Une réflexion sera menée sur les limites de l’application de modèles de langage naturel sur des données musicales, d’un point de vue technique comme d’un point de vue épistémologique, et sur les perspectives de modèles originaux spécifiquement adaptés à la modélisation de données musicales.

Plus de détails sur la page : http://www.algomus.fr/jobs/phd-nlp-en/

Profil du candidat :
Fort intérêt pour l’apprentissage automatique, le traitement automatique du langage naturel, et la musique.

Formation et compétences requises :
* Master d’informatique ou équivalent, apprentissage automatique, traitement automatique du langage naturel
* Connaissances et pratique musicales souhaitées.

Adresse d’emploi :
CRIStAL, Inria, Université de Lille, Villeneuve d’Ascq

Offre en lien avec l’Action/le Réseau : – — –/– — –

Laboratoire/Entreprise : Laboratoire Hubert Curien, University of Saint-Eti
Durée : 3 years
Contact : amaury.habrard@univ-st-etienne.fr
Date limite de publication : 2022-05-09

Contexte :
A Phd position covering Deep Learning, color Science, Photonics and Laser Processing is available at the University of Saint-Etienne

Sujet :
Laser processing is a flexible and cost-effective tool that recently opened new perspectives of applications in industry. Implemented on plasmonic metasurfaces, it produces very singular colors that can be tuned independently in different modes of observation. Laser-induced plasmonic colors thus enable printing multiplexed images, which have great promise in security printing and data storage1,2. However, the latter require very good accuracy in color printing. And, laser induced colors strongly depend on the initial state of the material. Predicting the full gamut of colors that can be observed in different modes of observation on plasmonic metasurfaces processed by a large set of laser processing parameters, when the initial state of these metasurfaces can vary from one batch to another, appears then as a crucial step for industrial implementation. As physical models are missing for such predictions, other approaches must be found.
Deep learning represents one of the most powerful family of models in machine learning when one has to make some predictions from data having some local structure such as images or surfaces. In this thesis, the objective is to provide some appropriate architectures accompanied with relevant objective functions to correctly train these architectures for accurate prediction of the laser printed colors3. A first challenge is to take into account physical properties of the materials and the laser processing parameters in the model. Then, another goal is to improve the robustness of the model by adapting existing adversarial robustness methods existing in image classification to laser printed colors4. Finally, the third aspect tackled in this project is to develop models for being able to automatically adapt the learned models to slightly different initial metasurfaces by means of transfer learning/domain adaptation strategies. This last objective intends to offer a certain tolerance to unwanted variations in the initial metasurface elaboration while maintaining a very good accuracy on the prediction of the laser printed colors.
References
(1) N. Destouches, et al. “Laser-empowered metasurfaces for white light image multiplexing”, Adv. Func.
Mater., 31, 2010430 (2021)
(2) N. Dalloz, et al. “Anti-counterfeiting white light printed image multiplexing by fast nanosecond laser
processing”, Adv. Mater., 34, 2104054 (2021)
(3) M. Raissi. “Deep hidden physics models: Deep learning of nonlinear partial differential equations”,
Journal of Machine Learning Research, 19(25):1–24, 2018.
(4) A. Shafahi, et al. “Adversarially robust transfer learning”, in International Conference on Learning
Representations (ICLR), 2020.

Applications must be sent to both contact email addresses as soon as possible and before May 1st, 2022. Application can be written in French or English.
All applications must contain:
-a CV, with a possible list of publications and conferences, and the CEFR level in English (except if university courses were taught in English)
-a short motivation letter explaining why you should be successful in this
research work
-Bachelor degree and transcripts
-Master transcripts (at least semester 1 and 2. Semester 3 if available)
-An example of project that you have carried out in deep learning
-References of academics to be contacted (or recommendation letters).

Application deadline: May 1st, 2022

Start of the position: October 1st, 2022

Contacts: Prof. Nathalie Destouches: nathalie.destouches@univ-st-etienne.fr, and Prof. Amaury Habrard: amaury.habrard@univ-st-etienne.fr

Link:
https://laboratoirehubertcurien.univ-st-etienne.fr/en/teams/functional-materials-and-surfaces/job-opportunities/phd-offer-prediction-of-multidimensional-colors-printed-by-laser-on-plasmonic-metamaterials-using-deep-learning-and-adaptive-strategies.html

Profil du candidat :
Master of Science in Computer Science or in Physics

Formation et compétences requises :
-Very good knowledge and experience in machine learning and deep learning
-Background in color science and image processing
-Some knowledge in photonics and eager to expand his/her experimental skills in this field
-Open-minded, curious and interested in working with both computer scientists and physicists
-Ability to take initiatives and work in autonomy

Adresse d’emploi :
Campus manufacture
Laboratoire Hubert Curien, University of Saint-Etienne
Saint-Etienne, France

Document attaché : 202203101326_PhD offer 2022.pdf