PhD opportunities

A Dataflow language for the automatisation and verification of ML systems

Thesis proposal

Area of expertiseReal-time computer science, robotics, systems and control - Fontainebleau
Doctoral SchoolSystems Engineering, Materials, Mechanics, Energy
SupervisorM. Olivier HERMANT
Research unitMathematics and Systems
ContactHERMANT Olivier
Starting dateOctober 1st 2019
KeywordsAI, data flow
AbstractAmaris Lab develops an automation platform for data acquisition and processing processes. Processes are expressed in a workflow-oriented domain-specific language (DSL). This language is the backbone of the system that makes it possible to define and execute sequences of actions such as:
- Automated data mining
- Robot Process Automation (RPA)
- Extraction of information
- Data formatting (standardization, preprocessing, Feature Selection)
- Training and application of machine learning algorithms
A graphical interface allowing to create, visualize and control these workflows has been developed. The growth of the use of the platform requires extensions of the language in order to apply it to other domains. It also generates robustness, security, optimization, and workflow performance requirements
ProfileThe candidate must hold a Master 2 or an engineering degree with a specialty in computer science by September 2019 at the latest and possess good software engineering skills. Knowledge in language design and machine learning would be appreciated.
FundingConvention CIFRE
PartnershipLe doctorant rejoindra le laboratoire de recherche du ple Innovation de lentreprise Amaris et travaillera en troite collaboration avec le centre de recherche en informatique de MINES ParisTech.

Laboratoire Amaris
http:/lab.amaris.com
Les Templiers 950 route des Colles 06410 Biot FRANCE