Post Doc - Symbolic learning for processes optimization

Country: France;

City: Ghent

Vacancy added: 11.01.2019

Employer: CEA Tech

Vacancy type: PostDoc vacancy;

 

RESEARCH FIELD

Computer science and software

ABSTRACT

In the context of a R&D platform on symbolic artificial intelligence (knowledge-based systems), we would like to develop machine-learning capacities. One of the interests of such systems is to be able to explain their decisions, in line with GDPR which is applicable from May 2018. After a bibliographic study, the candidate will propose and develop evolution of existing methods or a new one for rule induction in presence of numerical or symbolic (categorical) variables. These algorithms will be applied to the prediction of the link between physical parameters of material elaboration (temperature, pressure, etc.) and desired technical features (such as resistivity). Results obtained will be compared with other classical techniques of machine learning. Furthermore, inducted rules will be evaluated by experts in material elaboration. We will be able to answer to the question: Does AI approach generate new knowledge?Two other laboratories will provide data and will make experiments during the project. They will be available for describing their knowledge. Thus, machine-learning and knowledge-modeling approaches will be compared.

LOCATION

DM2I (LIST)

Laboratoire d'Analyse des Données et d'Intelligence des Systèmes

Saclay

CONTACT PERSON

BOUDET Laurence

CEA

DRT/DM2I//LADIS

CEA SaclayDigiteo SaclayBat 565, PC 19291191 Gif-sur-Yvette Cedex

Phone number: 01 69 08 88 82

Email: laurence.boudet@cea.fr

START DATE

As soon as possible

« The age limit is 26 years for PhD offers and 30 years old for post-doc offers. »

 
Where to send resume: