Intérim PostDoc position in distributed learning and multi-agent systems
Description de L'emploi
Université catholique de Louvain
Ville: Louvain-la-Neuve WBR BE
PostDoc position in distributed learning and multi-agent systems
Written by P. Chatelain and J. Hendrickx
A Post-doctoral position is open in distributed learning and multi-agent systems in a collaboration between the Institute of mechanics, materials and civil engineering (IMMC) and the Institute of information and communication technologies, electronics and applied mathematics (ICTEAM) at UCLouvain (Belgium). This 2-year position is offered in the context of the ERC consolidator Grant WakeOpColl on Learning and Collective Intelligence for Optimized Operations in Wake Flows , at UCLouvain.
This position is to be filled from September 1, 2017. The position is fully funded and will involve a strongly interdisciplinary work between the involved research groups at UCLouvain (Aerodynamics and Control). The researcher will be supervised by Profs. Philippe Chatelain and Julien Hendrickx. Interactions (including scientific stays) with the international collaborating institutions: ETH Zurich, U Illinois at Urbana Champaign, and UCLA, will be encouraged.
The WakeOpColl project (Consolidator ERC grant, PI: P. Chatelain) proposes an Artificial Intelligence and bio-inspired paradigm for the control of multiple flow devices subjected to wake effects, such as airplanes or wind turbines. Physics dictate that every flow device leaves a wake, which can then impact negatively or favorably other devices downstream. Wake turbulence between aircraft in air traffic and wake losses within wind farms are prime examples of this phenomenon, and incidentally constitute pivotal challenges to their respective fields of transportation and wind energy.
In the project, we associate to each flow device an intelligent agent that pursues given goals of efficiency or turbulence alleviation. Every one of these flow agents now relies on machine-learning tools to learn how to make the right decision when confronted with wake or turbulent flow structures. At a system level, we employ Multi-Agent System and Distributed Learning paradigms . Based on Game Theory, we build a system of interactions that incite the emergence of collaborative behaviors between the agents and achieve global optimized operation among the devices.
JOB DESCRIPTION AND PROFILE REQUIREMENTS
The successful candidate will contribute to the development of control schemes at agent and system levels. More specifically, he/she will handle tasks related to (i) the application of learning techniques for agent control (ii) the development of collaborative schemes in a multi-agent system with flow-mediated interactions. He/she will interact with researchers closer to the aeronautical and wind energy applications, in particular on the coupling of the control schemes with the computational fluid dynamics tools.
We are seeking candidates with a PhD degree in a relevant domain of engineering (or equivalent) with experience in control and multi-agent systems and/or learning. Good programming skills and/or experience in modelling are a plus. Good communication skills are further required in this project, due to its collaborative nature. Due to the scope of the project, scientific leadership and initiative are expected from applicants to this position.
WHAT WE OFFER
- a highly dynamic and multidisciplinary working environment in an international university town
- a challenging and high-visibility research topic that involves intensive national and international collaborations
- an implication in world-leading research groups in control and computational fluid dynamics and an access to state-of-the-art modelling tools
- competitive salary and social benefits
HOW TO APPLY
Your application should be sent by e-mail to firstname.lastname@example.org , with “PostDoc position in distributed learning and multi-agent systems” as subject. Application should include a motivation letter, full CV, list of publications, academic transcripts, and contact information of two references. Short-listed candidates will be contacted for a teleconference or physical interview.
Institute of mechanics, materials and civil engineering: www.uclouvain.be/immc
Institute of information and communication technologies, electronics and applied mathematics: www.uclouvain.be/icteam
WakeOpColl project: https://sites.uclouvain.be/wakeopcoll/
Web site for additional job details
- REQUIRED LANGUAGES