Post-doc offer in Machine Learning

Fastening multiview boosting approaches using multi-armed bandits

Aix-Marseille University

Marseille, France

Expected Starting date: as soon as possible
Duration : 12 months

Keywords

Machine learning, Multimodality, Ensemble based methods, Multi-armed bandits

Context

This position is proposed in the context of the french project LIVES¹ for Learning with Interacting ViEwS: theoretical framework, algorithms, and benchmarks funded by the ANR - the french national research agency.

This project gathers 4 research labs: the Laboratoire Hubert Curien (LabHC) of Saint-Etienne, the Laboratoire d'Informatique de Paris 6 (LIP6), the Laboratoire d'Informatique et Systèmes (LIS, Marseille), and the Institut des Neurosciences de la Timone (INT, Marseille). It has for objective the development of well-founded machine learning frameworks for learning in the presence of multiple and interacting views and its application to real-world problems. It covers a large spectrum from learning theory, algorithms to real applications in neuro-imaging.

The Post-doc candidate will join the Machine Learning research team Qarma² of the Computer Lab. of Marseille LIS (France), and will have the opportunity to collaborate with the other partners of the project. The Qarma team gathers about 20 researchers and phD candidates, who focus on machine learning, with a strong theoretical background applied to biology and neurosciences.

SCIENTIFIC OBJECTIVES

The hired postdoc will ideally focus his research on the design of theoretical-based algorithms that scale up to the combinatorics inherent in the multi-view learning setting.

Research topics of interest include but are not limited to:

• greedy methods

• ensemble based learning algorithm

• multi-armed bandits

• development of new algorithmic approaches,

• dealing with big data

Required skills

We are searching for an outstanding and highly motivated candidate with:

• A PhD degree in machine learning.

• A list of excellent publications (at reference journals and conferences in machine learning and/or computer vision and/or NLP) that demonstrates the expertise of the candidate.

• Strong programming skills in Python or C.

• Fluent in either french or english, good written capabilities

Working environment

Marseille is the second town of France, by the Mediterranean sea. It holds the biggest french university – Aix-Marseille University. The Qarma research theme is member of the Data Science Pole of the Laboratoire d’Informatique et des Systèmes³: the team is located at Technopole de Chateau-Gombert⁴, north-east of Marseille, which gathers many companies, start-ups and research labs.

The successful candidate will have the opportunity to work in synergy with Ph.D. students, and manage internships if willing to. The expected salary is about 2100€ per month after taxes, including basic health care. Teaching opportunities can raise the base salary.

Application instructions

The candidates should send:

• a motivation letter explaining the relevance of the candidate for the position and including some preferred research topics on the subject

• a CV with a detailed list of publications

• names and contact details of at least two references. Applications will be accepted from today until the position is filled. Important notice: the administrative constraints for hiring new fellows requires 40 days of procedure before the position officialy starts.

• a interview will be organized as soon as possible – possibly through visioconference – during which the candidate will present herself in 10 minutes, and will comment one of the three papers below in 10 minutes.

https://www.lri.fr/~kegl/research/PDFs/BuKe10.pdf

http://jmlr.org/papers/volume15/dubout14a/dubout14a.pdf

http://cell.missouri.edu/media/publications/Peng-TNNLS-2017.pdf

Contact information, to who the application must be sent

Cécile Capponi, Associate Professor, AMU. Cecile.Capponi@lis-lab.fr