4-year Postdoc position: Machine Learning for Graph-based High-dimensional Data Modeling

The Viper group at the University of Geneva (Department of Computer Science, Faculty of Sciences), headed by Prof. S. Marchand-Maillet, in collaboration with the DMML group of Prof A. Kalousis opens a 4-year Postdoc position in the field of High-dimensional Data Modeling based on graph modeling. A specific domain of application is proposed to be Flow Cytometry data modeling and analysis, in collaboration with existing efforts in the group, jointly with Geneva Hospital (see these introduction videos: [ intro, basic analysis ]).

Modeling high-dimensional data requires to defeating the statistical adversity of the curse of dimensionality. Such a modeling relies on the definition of relevant neighborhoods and their thorough exploitation. In such a context, (neighborhood) graphs and their related theoretical apparatus are the tools of choice for the analysis. The aim is to propose solutions to the fundamental problem of high-dimensional data modeling that will impact all fields of Data Science and Machine Learning.

Applicants must hold a recent PhD in Science and demonstrate a strong background in mathematics of high-dimensional spaces (linear algebra, optimization, probabilities, statistics,… ) and graph theory (spectral graph theory, random walks,…) along with practical capabilities (data management, programming, …). They should show an interest in collaborating with practitioners (medical professionals), work with real data and support PhD students in their research.

The postdoc position is an opportunity to develop one's own research and also to collaborate with other members (PhD, Master's) in the group.

Starting date: as soon as possible.

Applications must be composed at least of a resume clearly including a list of past achievements of the applicant (possibly the PDF thesis if in English or French), a letter of motivation and a short (eg 1-page) research statement in relation to the topic of the position offered (all in PDF format, grouped in *one unique* compressed file whose name starts 220402) and posted at the following link:

http://viper.unige.ch/apply

along with an email signaling your application and the name of the posted compressed file and the position applied to (for obvious reasons, any other form of application or incomplete/irrelevant applications will NOT be considered nor notified).

The University of Geneva is one of the oldest and the third largest university in Switzerland by number of students. It is highly ranked, both Nationally and Internationally according to the Shanghai Ranking of World Universities 2020.

The city of Geneva is very international and occupies a central position in Europe, and has a strong international reputation, including via the light of the international organizations (e.g. UN, WHO, WIPO, WTO, ICRC, CERN). The official language is French and the successful candidate may be asked from 2nd year on to interact as Teaching Assistant with French-speaking students.

Up-to-date information at http://viper.unige.ch/jobs

2204.pdoc.txt · Last modified: 2022/07/19 14:41 by marchand
--

Keywords: machine learning, information geometry, data mining, Big Data, affective information retrieval (recherche d'information), information visualisation, content-based image and video retrieval (CBIR, CBR, CBVR, CBMR, CBMIR), information mining, classification, multimedia and multimodal information management, semantic web, knowledge base (RDF, OWL, XML, metadata, auto-annotation, description), multimodal information fusion