Enjeu: End of fossil resources and ecological problems make energy management harder. TAO has various contacts and collaborations around energy (in industry and academia), so that works around that have an impact. Scientific challenges of parallel policy optimization are crucial for these works.
Technically: The problem is the optimization of investment policy, with uncertainties. Therefore, it's about Partially Observable Markov Decision Processes (POMDP). Includes security elements, parallelization, understanding of artificial intelligence aspects.
Context
The TAO team (http://tao.lri.fr) has several works in progress around long-term planning, in particular for the application to investment policies for energy management (horizon 2050; in Europe and North-Africa, also preliminary works in Taiwan).
Objectives
New parallel methodologies for optimizing policies.
Application to investment policies in energy management.
Useful competences:
- general knowledge in optimization and/or machine learning
- general knowledge in parallelism (message-passing, shared memory,
multicore machines, clusters, grids)
- Linux, C, C++, MPI, OpenMP and/or multithreading
(one can apply without all these competences of course)
Détails
Expected funding
Institutional funding
Status of funding
Expected
Candidates
Utilisateur
olivier.teytaud
Créé
Mercredi 18 avril 2012 07:11:01 CEST
dernière modif.
Mercredi 18 avril 2012 07:11:01 CEST
Fichiers joints
filename
créé
hits
filesize
Aucun fichier joint à cette fiche
Connexion
Ecole Doctorale Informatique Paris-Sud
Directrice
Nicole Bidoit Assistante
Stéphanie Druetta Conseiller aux thèses
Dominique Gouyou-Beauchamps
ED 427 - Université Paris-Sud
UFR Sciences Orsay
Bat 650 - aile nord - 417
Tel : 01 69 15 63 19
Fax : 01 69 15 63 87
courriel: ed-info à lri.fr