CANCELLATION OF THE EVENT DUE TO THE CORONAVIRUS SITUATION.
OFFICIAL RECTORAL ORDER FOR CLOSURE OF UNIVERSIDADE DO MINHO – click
We have the regret to announce that the workshop Machine Learning for CFD is postponed to 2021.
The accepted talks could be maintained but the registration procedure is discarded.
We plan to gather informally on Wednesday and Thursday evenings for diners in classical Portuguese restaurants.
Should you desire to participate to any of them?
Please let us know as soon as possible by email at:
Machine learning (ML) is among the Artificial Intelligence technologies with the greatest promise for CFD computation.
ML uses a set of training data to teach a computer program to achieve predictive capabilities they are not explicitly programmed to do.
However, machine learning takes time and massive amounts of data and remain obscure mathematical objects.
Artificial neural networks (ANNs) are making ML more effective by building an architecture of neurons with layers leading to so-called deep-learning strategy.
Researchers have applied deep learning to the complex task of computational fluid dynamics (CFD) simulations.
Because ANN can learn complex dependencies between high-dimensional variables and events, this may become and appealing technology for researchers who take a data-driven approach to CFD.
This field is under scrutiny by the community of CFD and this workshop aims at gathering researchers involved in numerical scheme development and numerical simulation of complex phenomena.
Talks and long session of discussions will be schedules during these intense two days in Portugal.
Some non-exhaustive list of thematics treated during the workshop would be:
- Use of Neural Networks to improve numerical methods;
- (un)supervised reinforcement learning to enhance numerical scheme capability;
- ML to replace AMR indicator, preconditioner for ill-posed systems, improve NL solver (initial guess, …);
- Active flow control via ML;
- etc.
Registration
The participation is free but participants should register with no exception.
before February the 1st
To register or submitt an abstract you can fill the forms online or send an email to:
Organization Comittee
S. Clain, J. Figueiredo, D. Lopes, G. Machado, T. Malheiro, R. Pereira
Center of physics, School of Sciences,
Universidade do Minho, PT
R. Loubère
Institut de Mathématiques de Bordeaux,
Université de Bordeaux, FR
Registration
The participation is free but participants should register with no exception.
before February the 1st
To register or submitt an abstract you can fill the forms online or send an email to:
Access and Hotels
The closest airport is the international Porto airport easily accessible by metro, bus, train or car.
The area of Guimarães is packed with hotels but we urge the participants to book well in advance as the touristic season never ends.
We are looking forward to welcoming you in the magnificient Minho region in Portugal.
Access and Hotels
The closest airport is the international Porto airport easily accessible by metro, bus, train or car.
The area of Guimarães is packed with hotels but we urge the participants to book well in advance as the touristic season never ends.
We are looking forward to welcoming you in the magnificient Minho region in Portugal.
Seminar room:
Departamento de matemática, Escola de Ciências, building Nº 12
Campus de azurem, Universidade do Minho, 4800-058 Guimarães
GPS: 41.452663 N, 8.289715 W
Organization Comittee
R. Loubère – Université de Bordeaux, FR
S. Clain, J. Figueiredo, D. Lopes, G. Machado, T. Malheiro, R. Pereira – Universidade do Minho, PT
Seminar room:
Departamento de matemática, Escola de Ciências, building Nº 12
Campus de azurem, Universidade do Minho, 4800-058 Guimarães
GPS: 41.452663 N, 8.289715 W