Dieses Dokuwiki verwendet ein von Anymorphic Webdesign erstelltes Thema.

Job offers

Permanent links


Le Cerfacs ouvre un poste d’ingénieur HPC pour renforcer ses actions dans le domaine. Le poste est à pourvoir pour 12 mois renouvelable à Toulouse. Vous pouvez postuler en nous contactant par mail monnier@cerfacs.fr. Niveau Minimum Requis : PhD ou Ingénieur Contacts : Nicolas Monnier, DSI, monnier@cerfacs.fr - Isabelle d’Ast, dast@cerfacs.fr & Gabriel Staffelbach, Gabriel.Staffelbach@cerfacs.fr. Ingénieurs HPC. Dates et lieu : Poste à pourvoir au Cerfacs à Toulouse à partir de décembre 2017 pour une durée de 12 mois.

2-year postdoc position within project FACTORY (New paradigms for latent factor estimation), funded by the European Research Council under a Consolidator Grant. Full announcement available at http://projectfactory.irit.fr/announcement.pdf

September 2017

PhD Position at National Centre for Meteorological Research, Toulouse, France

Title : Object processing of convective-scale model outputs

Supervisors : Dr Laure Raynaud, Dr Philippe Arbogast, Dr Etienne Mémin (HDR)

laure.raynaud@meteo.fr, philippe.arbogast@meteo.fr, etienne.memin@inria.fr


The French convective-scale Arome model, operational at Météo-France, is able to accurately represent some severe weather events, such as thundertsorms, heavy precipitation, fog or strong winds. However, the first years of Arome utilization suggest that these forecasts are affected by position, amplitude and timing errors. In order to improve these deterministic forecasts, an ensemble prediction system based on the Arome model has recently been developed, and provides an estimation of the forecast uncertainty. The development of relevant post-processing methods is another way of improving forecasts. Among them, a possible solution is the object-based approach: the main idea behind object-oriented processing consists in extracting the predictable signal from forecasts, under the form of coherent features, while the smaller and less predictable scales are filtered out. In this context, Arbogast et al. (2016) and Destouches (2017) proposed a probabilistic approach to automatically detect and track precipitation objects. The method is based on the use of segmentation methods for the detection part and of a stochastic particle filter for the tracking part. The goal of the PhD is to pursue this work and to extend its application to other meteorological parameters such as cloud cover and wind gusts.

Organization of the PhD

The first part of the PhD plans to refine the current detection/tracking algorithm for precipitation forecasts. In particular, the goal is to provide a robust algorithm, able to automatically detect and recognize precipitation of different types.

The verification of object detection/tracking methods has been mainly subjective so far. Objective verification will be the next important aspect to consider, in order to quantify the added value of the object processing.

Finally, this work will be extended to other weather parameters. Basically, all parameters with a high spatial and/or temporal degree of intermittency could benefit from this object processing.

Good knowledge of numerical modelization, data assimilation and image processing would be useful.


M. Destouches, 2017 : Detection and tracking of precipitation objects in convective-scale forecasts, Research internship report. Arbogast, P., O. Pannekoucke, L. Raynaud, R. Lalanne and E. Mémin, 2016 : Object‐oriented processing of CRM precipitation forecasts by stochastic filtering. Quart. J. Roy. Meteor. Soc. Raynaud, L., and F. Bouttier, 2016: Comparison of initial perturbation methods for ensemble prediction at convective scale. Quart. J. Roy. Meteor. Soc., 142, 854-866.

July 2017

PhD Position at LIPhy, Grenoble Alpes University (France) and College of Engineering, Swansea University (UK) (Physics/Applied Mathematics).

Studying the visco-elastic behaviour of lymph through experimental and computational micro-fluidics

PhD Position at Troyes: Aging with covariates, estimation and prediction

Thèse à Troyes : POlitiques de Maintenance adaptatives pour un système Multi-composants évoluant dans un Environnement Stressant

PhD position at LIRIS and Data R&D Institute at EMLyon Business School

PhD Proposal : String embeddings for large-scale machine learning in genomics


The cost of DNA sequencing has been divided by 100,000 in the last 10 years [1]. It is now so cheap that it has quickly become a routine technique to characterize the genomic content of biological samples with numerous applications in health [2], food or energy [3]. The output of a typical DNA sequencing experiment is a set of billions of short sequences, called reads, of lengths 100~300 in the {A,C,G,T} alphabet ; these billions of reads are then automatically processed and analyzed by computers to get some biological information such as the presence of particular bacterial species in a sample, or of a specific mutation in a cancer.

As the throughput of DNA sequencing continues to increase at a fast rate, the major bottleneck in many applications involving DNA sequencing is quickly becoming computational. The goal of this PhD project is to advance the state-of-the-art and propose new solutions for storing and analyzing efficiently the billions of reads produced by each experiment.

More precisely, we will focus on two important applications of DNA sequencing : - metagenomics, where the goal is to assign each read to a bacterial species in order to quantify the species that may be present in the sample analyzed ; - RNA-seq, where the goal is to assign each read to a gene, in order to quantify the level of expression of all genes in the sample analyzed. The basic problem to be solved in both applications is to assign each read to one among a set of known, longer target sequences (bacterial genomes or gene sequences). Standard techniques to solve that problem try to align each read to each target, using tools such as BLAST [4], BWA [5] or BOWTIE [6]. However, the computational cost of these techniques becomes prohibitive with current large sequence datasets, and faster alternative have been proposed recently. In particular, the problem can be reformulated as a supervised multiclass classification problem and solved by machine learning techniques such as naive Bayes [7] or support vector machines (SVM) [8]. We recently showed that large-scale machine learning techniques are competitive in accuracy and much better in computational cost that alignment-based methods for metagenomics applications [9].

The standard approach to solve the machine learning formulation is to represent each read as a fixed-length vector and then to train a linear classifier. A typical representation is to count the number of occurrences of each k-mer in a read, and to store these counts in a 4^k – dimensional vector, where k is an integer between 8 and 15. Recently, different representations using gapped k-mers and locality-sensitivity hashing (LSH) have been proposed and led to promising results [10], suggesting that there exists room for improvement in the way we represent reads as vectors for large-scale machine learning.

In this context the PhD candidate will investigate and propose new ways to represent DNA sequencing reads, that would lead to both (i) a compact representation for efficient storage and fast processing, and (ii) good performance in read classification for metagenomics and RNA-seq applications. Techniques to be investigated will include, in particular : - Random features to approximate string kernels [11,12] - LSH-based representations, including minHash [13] - Deep learning-based representations, including convolutional [14] and recurrent neural networks


PhD supervised by Jean-Philippe Vert (MINES ParisTech / Institut Curie / ENS Paris) : http://members.cbio.mines-paristech.fr/~jvert/

PhD fellowship of MINES ParisTech (about 1,690 €/month, net salary)

The ideal candidate should have a background in statistical machine learning, and a keen interest in biological applications (but no prior background in biology is needed).

To apply : send CV, transcripts and contact informations of two persons I could reach for recommendation to jean-philippe.vert@mines-paristech.fr before July 6, 2017.

Do not hesitate to reach him by email if you have any question.


[1] https://www.genome.gov/sequencingcostsdata/
[2] https://cancergenome.nih.gov
[3] http://www.hydrocarbonmetagenomics.com
[4] Altschul, S. F., Gish, W., Miller, W., Myers, E. W., & Lipman, D. J. (1990). Basic local alignment search tool. Journal of molecular biology, 215(3), 403-410.
[5] Li, H. (2013). Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. arXiv preprint arXiv:1303.3997.
[6] Langmead, B., & Salzberg, S. L. (2012). Fast gapped-read alignment with Bowtie 2. Nature methods, 9(4), 357-359.
[7] Qiong Wang, George M Garrity, James M Tiedje, and James R Cole (2007). Naive bayesian classifier for rapid assignment of rrna sequences into the new bacterial taxonomy. Applied and environmental microbiology, 73(16):5261–5267.
[8] Kaustubh R Patil, Peter Haider, Phillip B Pope, Peter J Turnbaugh, Mark Morrison, Tobias Scheffer, and Alice C McHardy (2011). Taxonomic metagenome sequence assignment with structured output models. Nature methods, 8(3):191–192.
[9] Vervier K., Mahé, P.,Tournoud, M., Veyrieras, J.-B., and Vert, J.-P. (2016). Large-scale machine learning for metagenomics sequence classification. Bioinformatics, 32(7) :1023-1032.
[10] Luo, Y., Yu, Y, Zeng, J., Berger, B. and Peng, J. (2017) Metagenomic binning through low density hashing. biRxiv 133116.
[11] Rahimi, and Recht, B. (2007) Random features for large-scale machine learning. In NIPS 2007.
[12] Mourragui, S. (2017). Random projections for large-scale metagenomics classification. Internship report, MINES ParisTech.
[13] Indyk, P. and Motwani, R. (1998). Approximate nearest neighbor: Towards removing the curse of dimensionality. In Proceedings of the Symposium on Theory of Computing.
[14] Zhang, X., Zhao, J., and LeCun, Y. (2016). Character-level Convolutional Networks for Text Classification. arXiv 1509:01626.

Société Orange : Proposition de sujets de thèse

Expéditeur: regine.angoujard@orange.com Date: 13 juin 2017 à 08:35:23 UTC+2 Destinataire: FAINÉANT Virginie IMT/OLR virginie.faineant@orange.com Cc: ANGOUJARD MIET Régine IMT/OLR regine.angoujard@orange.com Objet: Orange, votre partenaire pour les thèses de vos étudiant(e)s


Nous avons le plaisir de vous informer que nous avons sélectionné votre établissement pour proposer nos sujets de thèses 2017 en priorité à vos étudiant(e)s.

Depuis de nombreuses années, Orange recrute une quarantaine de doctorants sur des sujets touchant à la fois aux sciences et techniques relatives aux réseaux, aux plateformes de services, aux services et aux usages. Nous souhaitons renforcer la proportion de doctorants issus des cursus de recrutement prioritaires pour Orange, et diversifier les origines thématiques, en ne nous limitant pas exclusivement aux formations en Télécoms, mais en incluant notamment les domaines de l’énergie, des systèmes, de l’informatique et des mathématiques.

Vous trouverez, en avant-première, la liste des sujets de thèses que nous avons sélectionnés pour le programme doctoral 2017.

Nous vous demandons donc de faire la promotion de ces sujets auprès de vos étudiants, sur la base de l’argumentaire suivant :

Orange considère ses doctorants comme des acteurs clés du succès de sa recherche. Ils lui apportent des connaissances scientifiques ou technologiques et des pistes d’innovation nouvelles, considérées comme une priorité stratégique par le Groupe.

Pendant leurs trois années de thèse, nos doctorants, recrutés en CDD, participent avec leurs encadrants à une grande variété de projets, internes au Groupe ou coopératifs (ANR, projets européens…). Ils peuvent bénéficier d’une formation spécifique de préparation à la vie active financée par le Groupe auprès de l’ABG (Association Bernard Gregory).

La politique doctorale du Groupe permet à nos doctorants de construire des liens et des synergies efficaces avec le monde de la recherche publique (Institut Mines-Télécom, INRIA …), et aussi d’avoir une première expérience très significative du travail en entreprise, ce qui ouvre des perspectives de carrière privilégiées à la fois dans le domaine de la Recherche (publique ou privée) et aussi dans les Entreprises, et pas uniquement dans leurs laboratoires de recherche.

Pour plus de détails, ces offres sont actuellement consultables sur Orange.jobs thèses , les candidats intéressés devront obligatoirement postuler en ligne.

Je me tiens à votre disposition pour tous renseignements complémentaires,


Régine Angoujard Miet Gestion du programme doctoral ORANGE/IMT/OLR/DOP/PRA Fixe : +33 1 57 39 93 31 regine.angoujard@orange.com

PhD Scholarship available (UNSW Sydney, Australia)

Deadline: 21 July, 2017

Details here: http://web.maths.unsw.edu.au/~lafaye/#opportunities

The UNSW Scientia Ph.D. Scholarship Scheme is the most prestigious and generous scholarship scheme at UNSW. It aims to attract the best and brightest people into strategic research areas. Awardees receive a $50,000 scholarship package for four years, comprising a $40,000 per annum tax-free stipend and a travel and development support package of up to $10,000 per annum. International students also receive a tuition fee scholarship. In addition to this scholarship package, scholars are provided with access to a range of development opportunities across research, teaching and learning and leadership and engagement.

The funded project aims to develop new tools and insights for insurer risk management by combining modern statistical learning (‘data analytics’, ‘big data’, ‘predictive analytics’) techniques with actuarial risk theory. The findings will allow for accurate and equitable rating and measurement of risks, and ultimately contribute to sustainable and equitable protection for policyholders. For equity and stability, insurers must be able to assess their risks accurately. Nowadays, they have access to an increasing number of data sources of very different types, and in finer and finer detail. This interdisciplinary project is concerned with 21st century estimation of insurance risks, and proposes to deal with all of the four V’s of big data: volume, velocity, variety and veracity. The focus will be on the extension of recent statistical analytics including in particular deep learning. Insights developed with this analysis will be further incorporated into concepts from actuarial risk theory.

The supervisory team will comprise Benjamin Avanzi and Bernard Wong (both Associate Professor at the Business School, Risk and Actuarial Studies) and Pierre Lafaye de Micheaux (Senior Lecturer, School of Mathematics and Statistics).

The candidate must have a strong background in statistics/mathematics and good programming skills (preferably in R and C/C++; some experience with Linux would be an asset).

If you are interested, please contact either of us for more details, joining a recent CV and a copy of your academic transcripts.

June 2017

Data Scientist at Inserm Orsay & Paris


  • Unordered List ItemAfin de promouvoir les compétences des ressources humaines des pays en développement et de favoriser la compréhension, l’amitié entre les nations et le peuple Français, la Commission Nationale Française (CNF) pour l'UNESCO en accord avec le Secrétariat d’Etat Français à l’Education (SEFE) met 100 (cent) bourses d’études à la disposition des étudiants étrangers désireux d’effectuer un séjour d’étude ou de recherche en France pour l’année académique 2017/2018.
  • Un formulaire de présélection a été conçu en PDF et mis à la disposition de tous les candidats afin de constituer le dossier de candidature. Merci de bien vouloir faire une demande dudit formulaire auprès Secrétariat d’Etat Français à l’Education (SEFE) en envoyant une lettre de motivation mentionnant votre pays d’origine afin d’être orienté pour le dépôt de votre dossier de candidature
  • Courriel du département de bourse d’étude : sefe.gov.fr@diplomats.com

Secrétariat d’Etat Français à l’Education (SEFE)

PhD proposal / Thèse CIFRE : data-science for chip manufacturing, at STMicroelectronics

  • STMicroelectronics (Crolles) et le Laboratoire G-SCOP (Grenoble) propose une thèse CIFRE :
  • Mission : Science des données pour l'amélioration des processus de conception et de fabrication des puces de silicium.
  • Profil recherché : compétences en analyse de données, statistiques et programmation; intérêt pour les applications industrielles et l'évolution dans un environnement complexe et pluridisciplinaire

Voir aussi http://st.mycvthequehq.com/offre-fr-e3c12b79d162.html

Contacts : pierre.lemaire@grenoble-inp.fr, bertrand.le-gratiet@st.com

7 prestigious PhD Student Positions with the REVOLVE project


PhD position: Privacy risk assessment and algorithms for matching and enriching personal and professional profiles across social networks

We may have a PhD subject to propose on “Privacy risk assessment and algorithms for matching and enriching personal and professional profiles across social networks” (see short description below (*)).

This PhD work would be done in collaboration between the SLIDE team of the LIG (Laboratoire d'Informatique de Grenoble) and the talent.io company in Paris, and funded by a CIFRE convention. The candidate would be employed by the company for 3 years to conduct his PhD work, co-supervised by Oana Goga and Marie-Christine Rousset (who are researchers at LIG).

Let us now if you are interested by the subject and the industrial setting of this PhD work. If this is the case, Nicolas Meunier CEO of the talent.io company will get in touch with you for further discussions.

(*) Short description:

talent.io is a linking platform for helping developers to find the best jobs for them according to their profile, and tech companies to find the best developers for their needs. The distinguishing point of this platform is to push candidate profiles to companies that have registered to the platform. It is therefore very important that the pushed profiles are the most complete and precise as possible so the right candidates are contacted by the right companies as fast as possible. At the moment, the profiles are mainly built manually, first by the candidates when they sign up and then enriched by talent.io staff based on phone interviews with candidates. For candidates providing their linkedin profile, some profile items are automatically extracted from their linkedin profile. The goal of the project is to make the profile construction more automatic by taking benefit of existing profiles that can be found across social or professional networks. In particular, talent.io has already collected 5.5 millions of github profiles. It is also registered to services that provide on demand social network profiles matching with a keyword query. The first step of the project will be dedicated to collect a sample of profiles data and to put them in an appropriate format for applying some of the state-of-the-art profile matching algorithms. The second step will consist in comparing the performance and the results of the different chosen profile matching algorithms on these profile data. In particular, the number of false positive matches is an important criteria for assesssing the quality of a profile matching algorithm. Finally, potential solutions for exploiting aggregated profiles should be proposed in order to choose the profiles to push or to recommend to target companies.

May 2017

Permanent Research Engineer position in medical image processing

Starting date : September 2017 (could be adapted to availability)

Application: see http://mathhatt.free.fr/posteIR.pdf

Contact Dr Catherine Cheze Le Rest (Catherine.cheze-le-rest@chu-poitiers.fr) for more information. Interested candidate should send curriculum vitae, complete list of past research topics and publications, a motivation letter, and two references to Dr Catherine Cheze Le Rest (Catherine.cheze-le-rest@chu-poitiers.fr) . Review of applications will begin immediately and continue until the position is filled.

PhD position at Université Paris-Sud

PhD position within Institut Mines Telecom (Telecom SudParis)

PhD Title : Proactive Mobility, Naming and Caching in future Complex 5G Network Services: Modeling, Simulation and Experimentation.

Keywords: ICN, IoT, 5G, complex graphs, proactive models

Profile and skills required

  • A Master's degree or an engineer Computer Science and /or Applied Mathematics.
  • Good understanding of the fundamental of and network science in general
  • Knowledge in optimization theory, machine learning theory, graph theory, stochastic processes, Bayesian networks and/or Game theory are highly desirable.
  • Programming potential in various languages (Python, C/C ++,Matlab, Java,…etc).
  • A good level of English.
  • A strong curiosity in research interdisciplinary.
  • English speaking and writing.

Director and supervisor: Prof. H. Afifi and Prof. H. Moungla


Télécom-SudParis / Université Paris Saclay

CNRS - Telecom SudParis

CEA Saclay Nano-Innov

Avenue de la Vauve - Bat 861

91191 Gif-sur-Yvette, France

The PhD details are on the website (link below) of the Paris-Saclay University, and applications will have to be sent to: hassine.moungla@telecom-sudparis.eu & hossam.afifi@telecom-sudparis.eu

And to submit on the site of adum of the Paris Saclay University before May 12, 2017.


The PhD description will be in the pdf file (EN version) with the same link at the bottom of the web page.

This position is part of Carnot 2017 call, for October 2017.

PhD position at LMDC/IMT:

3-Year PhD position at IRSTEA:

3-Year PhD position in Saint-Etienne:

Probabilistic study of instantiated gaussian processes and application to spatio-temporal data.

Starting date: September or October 2017 Application deadline date: April 14th 2017

Decision announcement date: June 1st 2015


The thesis will take place in the Saint-Etienne part of Camille Jordan Institute. The research will be undertaken in the context of an interdisciplinary project involving also Hubert Curien Laboratory from the University Jean Monnet of St Etienne.

The consortium has scientific expertise on probability and statistics, information and image processing, and machine learning, providing a stimulating scientific environment for this thesis. Last but not least, St Etienne is a very pleasant place to study and work. St Etienne is rated each year as one of the best place in France for studying.

PhD thesis subject

Gaussian processes are non-linear models of continuous random processes which are widely used to describe numerical data as sounds, images, videos, etc. (see for e.g. [W08,Z16]).

A Gaussian process is defined mainly by its expectation function and its covariance function (the kernel).

The description of the kernel using parametric functions and the estimation of these parameters form the focus of many recent works [L05,D16].

In the context of image sequences (knowing that our study is intended to address other types of data), the main objective is no longer to describe a Gaussian process but a set of Gaussian processes that can possess instances (Different temporal or spatial supports), with the aim to analyse videos with dynamic textures (lights, waves, clouds, fields of wheat …) taken from different angles for example.

The main objective of the thesis is to provide a precise mathematical framework for these instanciated Gaussian processes in order to be able to estimate the different parameters (instances, mathematical expectations and kernels' parameters).

First, the PhD student will be intended to make a state-of-the-art about the different kernels and their properties, mainly their stationarity in time and space in order to propose new kernels. The next step is to develop robust parameter estimation methods and to work on the automatic selection of the kernels. Then, the formalism of non-stationary and instanciated Gaussian processes will be developed, together with their numerical simulations. The last step concerns the mixture of instanciated Gaussian processes and their application to real data like videos.


We are looking for a motivated student holding a Master degree (on the 1st of September 2015) in the field of applied mathematics (probability, data analysis, estimation and optimization, …) or “computer science” (or “computer vision”) with strong skills in applied mathematics. A good background in software development (algorithmic, Matlab/Octave/Scilab or Python, …) is expected. Knowledges in image processing and machine learning would also be appreciated.


Net salary: around 1400 euros without teaching activities and around 1650 euros with teaching activities (64 hours per year).

Application process

Your application should include the following documents: - Letter of intent - Grades and ranking during Master 1 and Master 2 - Scientific CV - List of publications (if it exists of course) - Names of Referees (at least 2)

Contacts: - tugaut@math.cnrs.fr (http://tugaut.perso.math.cnrs.fr/accueil.html) - olivier.alata@univ-st-etienne.fr (http://perso.univ-st-etienne.fr/ao29170h/)


[D16] N. Durrande1, J. Hensman, M. Rattray, N. D. Lawrence, “Detecting periodicities with Gaussian processes.” PeerJ Computer Science 2:e50 https://doi.org/10.7717/peerj-cs.50. [L05] Neil Lawrence, “Probabilistic Non-linear Principal Component Analysis with Gaussian Process Latent Variable Models.” Journal of Machine Learning Research 6 (2005) 1783–1816. [W08] Jack M. Wang, David J. Fleet and Aaron Hertzmann, “Gaussian Process Dynamical Models for Human Motion.” IEEE Trans. On Pattern Analysis and Machine Intelligence, vol. 30, no. 2, Feb. 2008. [Z16] Ziqi Zhu, Xinge You, Shujian Yu, Jixin Zou and Haiquan Zhao, “Dynamic texture modeling and synthesis using multi-kernel Gaussian process dynamic model.” Signal Processing, Vol. 124, July 2016, Pages 63–71. Big Data Meets Multimedia Analytics — Containing a selection of papers from the 21st International Conference on Multimedia Modelling (MMM2015).

March 2017

C. Geuzaine and Xavier.Antoine@univ-lorraine.fr are proposing a joint Ph.D. thesis with on some numerical methods in optics (classical/quantum) in the framework of Inria. The potential position is located mainly in Nancy but through a close co supervision with the University of Liège. All the information is available at


where the students can apply.

IFP Energies nouvelles propose un stage en 2017 en régression parcimonieuse et réduction de dimension appliquées à la normalisation de mesures instrumentales et de données présentant un facteur d'échelle différent. L'objectif est de l'estimer en présence de données manquantes et aberrantes, et de bruits, pour des signaux courts, présentant des variations d'amplitude importantes. Cette estimation doit se faire de la manière la plus automatisée possible, en se basant sur les propriétés et des a priori sur les données (parcimonie, positivité)

Le sujet est décrit (en anglais), sur la page :


PhD Student position in Numerical Simulation at LJLL (Paris)

# Links for full description : https://www.ljll.math.upmc.fr/~privat/documents/sujet_these_LRGM.pdf

PhD student position in Machine Learning for Geosciences [ERC project]

We are searching for an outstanding candidate with a strong interest in machine learning and geosciences to cover one PhD student position to join the Image and Signal Processing (ISP) group in the Universitat de Valencia, Spain, http://isp.uv.es. The position is fully funded by an ERC Consolidator Grant 2015-2020 entitled “Statistical Learning for Earth Observation Data Analysis” (SEDAL), http://isp.uv.es/sedal.html, under the direction of Prof. Gustau Camps-Valls.

  • The project and job description

We aim to develop the next generation of statistical inference methods to analyze Earth Observation (EO) data. Machine learning models have helped to monitor land, oceans, and atmosphere through the analysis and estimation of climate and biophysical parameters. Current approaches, however, cannot deal efficiently with the particular characteristics of remote sensing data. We will develop advanced regression (retrieval, model inversion) methods to improve efficiency, prediction accuracy and uncertainties, encode physical knowledge about the problem, attain self-explanatory models, learn graphical causal models to explain the complex interactions between essential climate variables and observations, and discover hidden essential drivers and confounding factors in Climate/Geo Sciences.

Highly motivated researchers with a degree in computer science, statistics, machine learning, electrical engineering, physics, or mathematics are encouraged to apply. All candidates should have a solid understanding and knowledge of machine learning and statistics, and being particularly interested in remote sensing and geoscience problems. The thesis will address problems in regression, graphical models and causal inference. Good programming skills (Matlab/Python/R/C++), a critical and organized sense for data analysis, as well as maturity and commitment, strong communication, presentation and writing skills are a big plus.

  • Application details

- Deadline: Send your application no later than April 1st 2017. - How? Send me: 2-pages CV, motivation letter, papers if any, and one recommendation letter or contact - When? Preferred starting dates: June 2017 - How long? 3 years contract - How much? Salary according to UV scales including social security, health insurance benefits, and travel money - Where? Valencia, Spain, Mediterranean city, nice weather, hike and beach. Excellent cost-of-living index = 55

  • Contact

- Before applying: Informal inquiries may be addressed to Prof. Dr. Gustau Camps-Valls, gustau.camps@uv.es - Ready to apply? Send your dossier in one single PDF to gustau.camps@uv.es, subject: “SEDAL application”

Onera in collaboration with Telecom Paristech invites applications for a PhD studentship to undertake research in the fields of machine learning and remote sensing. Subject is “Deep networks for multi-temporal activity analysis of Earth-observation data”.

# Short description : Last years have seen the massive adoption of deep learning techniques for various tasks in computer vision. In remote sensing and Earth-observation data analysis, our team has developed algorithms for classification and detection which have established new state-of-the-art performances. With this new PhD thesis, we now want to discover how deep networks can help understanding the multitemporal satellite image series.

Research axis will include : * Semantic classification of aerial and satellite images * Deep Learning architectures * Investigating standard tools of image comparison in the context of deep network analysis. * Big data

# Links for full description : https://www.adum.fr/as/ed/voirproposition.pl?site=adumfr&matricule_prop=14566 http://sites.onera.fr/formationparlarecherche/theses-dtim (ref. TIS-DTIM-2017-008)

The successful candidates will work with : Pr. Yann Gousseau http://perso.telecom-paristech.fr/~gousseau/index_eng.html Alexandre Boulch https://sites.google.com/view/boulch/home Bertrand Le Saux http://www.onera.fr/en/staff/bertrand-le-saux at Onera ( http://www.onera.fr/en ) and Telecom Paris Tech ( https://ltci.telecom-paristech.fr/en/ ), located in Palaiseau, near Paris, France .

PhD Proposal - UQAM, Canada

The use of Determinantal point processes for Monte-Carlo experiments and computer experiments

Jean-François Coeurjolly & Pierre-Olivier Amblard

Objet: NPL-Cambridge CASE PhD Studentship 'Vegetation assessment using machine learning techniques on spectral imaging data' Date: 28 février 2017 à 20:07:37 UTC+1

NPL-Cambridge CASE PhD Studentship 'Vegetation assessment using machine learning techniques on spectral imaging data’

See also http://www.jobs.cam.ac.uk/job/13010/

It is estimated that by 2050 70% of more food needs to be produced worldwide. It is therefore not only essential to use our resources as efficiently as possible, but also to assess and mitigate the risk to crops. Farming is increasingly driven by machines and with less staff it is difficult for farmers to monitor their crop. There is the drive to use airborne technologies as well as satellite imagery to do this remotely and automate this. The project aims to build up an extensible, probabilistic framework to do this resulting in a database and data model.

The ultimate aim is to not just have a database of the spectral signatures of different plant species, but also to incorporate phenology and health status of the plants. With regards to crops this will help mitigate the risks of droughts and diseases. Irrigation can be directed where needed and fertilizer used more effectively. Early intervention can stop diseases spreading and there will be less use of pesticides and fungicides. Being able to choose an optimal time to harvest will lead to less food wastage.

Using sophisticated machine learning techniques, sparse models of the land cover can be created. These models will help where there is limited up and downlink bandwidth as there are with airborne technologies as well as with satellites. The device gathering the data can carry a sparse model and only where new data is significantly different to the model action is necessary. In the first instance this action will be an alert to an anomaly. Further analysis is then necessary whether the anomaly is expected due to e.g. change of the season, or the anomaly needs intervention or the model needs updating (e.g. a change of crop).

The aim of this PhD project is a probabilistic model of spectral signatures of plants incorporating phenology and diseases. To this end various machine learning techniques will be employed and benchmarked against each other. Data from the CropScape database (https://nassgeodata.gmu.edu/CropScape/) will be combined with data from Avaris (http://aviris.jpl.nasa.gov/alt_locator/) for initial analysis. This can then be enriched with data from the Sentinel satellites for temporal analysis, since the revisit times are shorter. Different resolutions and number of spectral bands need to be given consideration. The techniques will also be assessed on their ability to generate knowledge automatically, for example in which way the spectral signature of a plant changes under increasing drought conditions and whether there are underlying general principles. Another aspect is the development of a confidence measure identifying and quantifying classification error.

The successful candidate will be working closely with David Coomes (Cambridge), Anita Faul (Cambridge), Alistair Forbes (NPL), and Carola-Bibiane Schönlieb (Cambridge).

Applicants should have a masters degree in mathematics or a closely related subject, e.g. an engineering degree with a strong mathematical foundation, and should be UK or EU nationals.

The funds for this studentship are available for 3 years in the first instance.

In order to be considered for this studentship please submit a formal application to the PhD in Applied Mathematics and Theoretical Physics, University of Cambridge via the University's Graduate Admissions website (for more information on this please visit http://www.graduate.study.cam.ac.uk/how-do-i-apply); and send an expression of interest email to grad-administrator@maths.cam.ac.uk which explains why you are interested in this studentship.

Applications should be submitted online until the 31st of March 2017. Expressions of interest letter that briefly describe your motivation for this project should be sent to grad-administrator@maths.cam.ac.uk by the same date.

Please quote reference LE11533 on your application and in any correspondence about this vacancy.

The University values diversity and is committed to equality of opportunity. The Department would particularly welcome applications from women, since women are, and have historically been, underrepresented on our student cohort.

The University has a responsibility to ensure that all employees are eligible to live and work in the UK.

DEPARTMENT/LOCATION Department of Applied Mathematics and Theoretical Physics

REFERENCE LE11533 ; CATEGORY Studentships ; PUBLISHED 23 February 2017 ; CLOSING DATE

 31 March 2017 

February 2017

PHD Proposal

Title: Accelerating kinetic model construction and better process control by extracting information from transient data

Place: IFPEN, Rond point de l’échangeur, BP 3 - 69360 Solaize - France

PhD position at IFP Energies nouvelles (IFPEN)

The production of ultra-low sulfur diesel is a major challenge for economic and societal perspective. The hydrocracking process is dedicated to maximize the production of such pools. The use of kinetic models is mandatory in order to design and control such processes. These models must be recalibrated for each catalyst generation. This requires the use of long pilot plant tests (several months) and expensive (several hundred k€). The relatively long duration of the tests (1.5 to 2 months) is partly due to the catalyst stabilization at each operating conditions modification (several days) and secondly to the supervision of the test in order to reach targets (conversion efficiency). The objective of this thesis is multiple:

1. Optimize the operating conditions to maximize the information (design of experiments) and use the points in transient mode (points currently acquired to check the stability but not used) for kinetic models fitting

2. Optimize the supervision of pilot plants to reach quickly some targets (conversion, yields)

This will be carried out with following program:

1. Study experimental data in order to provide a catalyst stabilization mechanism

2. Numerical study in order to build the model, test and develop methods for on line identification and develop methods for design of experiments

3. Validation with experimental data

Desired Skills and Experience:

  • Degree in Process System Engineering, Data Mining/Data Science or Computer Science
  • Experience with one or more of the following: Programming skills, Process Engineering, Process Control
  • English skills
  • Motivation to work in a transdisciplinary team

Supervisors: Dr. Benoit Celse, Benoit.celse@ifpen.fr ; Dr. D. Guillaume, denis.guillaume@ifp.fr; Pr. J. Thybaut, Joris.Thybaut@UGent.be

Interested candidates are requested to send a detailed CV, one recommendation letter and university/master transcripts to Benoit Celse at benoit.celse@ifpen.fr.

Industrial PhD internship

The purpose of the internship is to developp a model to measure the economic power of the different agencies of the Bank La Caisse d'Epargne – Rhônes Alpes, that will take into account different external variables (INSEE indices, population transformation, density of rival agencies, …), the caracteristics of the agencies and consommation behaviors of the customers. This model should allocate the cost masses uniformly on the considered agencies.

This internship can create opportunities for further doctoral investigations (typically with a CIFRE funding).

Key-words : Generalized linear model, logistic regression, variable selection Academic supervisors : Cristina Butucea (CREST) and Clément Marteau (Université Lyon 1). Contact: marteau@math.univ-lyon1.fr

L'Observatoire de Paris offre un contrat de post-doc/ingénieur à durée déterminée d’un an, renouvelable plusieurs années, au laboratoire GEPI à Meudon pour travailler au développement du traitement de données spectroscopiques du satellite Gaia, mission de l’Agence Spatiale Européenne (ESA).

Une description plus détaillée du poste est disponible à l’adresse http://gaia.obspm.fr/qui-fait-quoi/article/postes-a-pourvoir-spectrocopie

December 2016


joboffers.txt · Last modified: 2017/10/15 17:05 by desbat
Dieses Dokuwiki verwendet ein von Anymorphic Webdesign erstelltes Thema.
www.chimeric.de Valid CSS Driven by DokuWiki do yourself a favour and use a real browser - get firefox!! Recent changes RSS feed Valid XHTML 1.0