Connectionists: Call for submission Special Issue Entropy Journal "Physics-Based Machine and Deep Learning for PDE Models" - Dead line 30 June 2022

[patrick Gallinari]

  Special Issue "Physics-Based Machine and Deep Learning for PDE Models"

https://www.mdpi.com/journal/entropy/special_issues/Physics-Based_Machine


Deadline for manuscript submissions:*30 June 2022*



<mailto:?&subject=From%20MDPI%3A%20%22Physics-Based%20Machine%20and%20Deep%20Learning%20for%20PDE%20Models%22&body=https://www.mdpi.com/si/82723%0A%0APhysics-Based%20Machine%20and%20Deep%20Learning%20for%20PDE%20ModelsMachine%20learning%20has%20been%20successfully%20used%20for%20over%20a%20decade%20for%20applications%20in%20engineering.%20It%20has%20recently%20started%20to%20attract%20attention%20for%20scientific%20computing%20in%20domains%20dominated%20up%20to%20now%20by%20the%20classical%20mechanistic%20modeling%20paradigm.%20It%20is%20particularly%20promising%20for%20domains%20involving%20complex%20processes%2C%20only%20partially%20known%20and%20understood%20or%20when%20existing%20solutions%20are%20computationally%20not%20feasible.%20This%20is%20the%20case%20for%20the%20modeling%20and%20simulation%20of%20complex%20dynamical%20physical%20systems.%20Classical%20modeling%20relies%20on%20PDEs%2C%20and%20simulation%20is%20central%20to%20engineering%20and%20physical%20science%20with%20applications%20in%20domains%20such%20as%20earth%20systems%2C%20biology%2C%20medicine%2C%20mechanics%20and%20robotics.%20Traditional%20simulation%20problems%20involve%20computational%20fluid%20dynamics%20and%20turbulence%20modeling%2C%20mechanistic%20design%20and%20many%20other%20domains.%20Such%20numerical%20models%20are%20also%20used%20intensively%20in%20industrial%20systems%20design%2C%20in%20simulation%20for%20decision%20support%2C%20or%20in%20safety%20studies.%20They%20are%20used%20for%20inversion%2C%20data%20assimilation%20and%20forecasting.%20Despite%20extensive%20developments%20and%20promising%20progress%2C%20this%20classical%20paradigm%20suffers%20from%20limitations.%20It%20is%20often%20impossible%20or%20too%20costly%20to%20carry%20out%20direct%20simulations%20at%20the%20scale%20required%20for%20natural%20or%20industrial%20problems.%20The%20physics%20may%20be%20too%20complex%20or%20unknown%2C%20leading%20to%20incomplete%20or[...]><https://twitter.com/intent/tweet?text=Physics-Based+Machine+and+Deep+Learning+for+PDE+Models&hashtags=mdpientropy&url=https%3A%2F%2Fwww.mdpi.com%2Fsi%2F82723&via=Entropy_MDPI><http://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.mdpi.com%2Fsi%2F82723&title=Physics-Based%20Machine%20and%20Deep%20Learning%20for%20PDE%20Models%26source%3Dhttps%3A%2F%2Fwww.mdpi.com%26summary%3DMachine%20learning%20has%20been%20successfully%20used%20for%20over%20a%20decade%20for%20applications%20in%20engineering.%20It%20has%20recently%20started%20to%20attract%20attention%20for%20scientific%20computing%20in%20domains%20dominated%20up%20to%20now%20by%20the%20classical%20mechanistic%20modeling%20paradigm.%20It%20is%20%5B...%5D><https://www.facebook.com/sharer.php?u=https://www.mdpi.com/si/82723>









Guest Editors
*Dr. Nicolas Bousquet*
1. CNRS, LPSM, Sorbonne University, 75005 Paris, France
2. EDF R&D, Industrial AI Lab SINCLAIR, Paris, France

*Prof. Patrick Gallinari*
1. CNRS, ISIR, Sorbonne University, 75005 Paris, France
2. Criteo AI Lab, 75009 Paris, France


    Special Issue Information

Machine learning has been successfully used for over a decade for 
applications in engineering. It has recently started to attract 
attention for scientific computing in domains dominated up to now by the 
classical mechanistic modeling paradigm. It is particularly promising 
for domains involving complex processes, only partially known and 
understood or when existing solutions are computationally not feasible. 
This is the case for the modeling and simulation of complex dynamical 
physical systems. Classical modeling relies on PDEs, and simulation is 
central to engineering and physical science with applications in domains 
such as earth systems, biology, medicine, mechanics and robotics. 
Traditional simulation problems involve computational fluid dynamics and 
turbulence modeling, mechanistic design and many other domains. Such 
numerical models are also used intensively in industrial systems design, 
in simulation for decision support, or in safety studies. They are used 
for inversion, data assimilation and forecasting. Despite extensive 
developments and promising progress, this classical paradigm suffers 
from limitations. It is often impossible or too costly to carry out 
direct simulations at the scale required for natural or industrial 
problems. The physics may be too complex or unknown, leading to 
incomplete or inaccurate models.

The availability of increasingly large amounts of data, either from 
observations or from simulations, and the successes witnessed by ML 
methods on large size or large dimensional problems has opened the way 
for exploring the data driven modeling of complex dynamical physical 
phenomena. ML based techniques may accelerate simulations, acting, for 
example, as reduced models. More generally, a promising direction 
consists in integrating physics-based models with machine learning. This 
raises several challenges such as how to perform such decompositions, 
how to train such combined systems, how to handle discretization errors 
or guarantee numerical stability of the solutions, how to handle 
out-of-sample scenarios, and how to ensure physical consistency of the 
solutions.

An additional challenge is the shift from academic case studies to 
realistic problems representing complex phenomena. Current solutions are 
most often demonstrated on simulated problems and there is still a large 
gap between academic and real-world developments.

This Special Issue, therefore, aims to gather specialists from different 
disciplines and to enable the dissemination of their recent research at 
the crossroad of model based and data based dynamical physical system 
modeling and on “physically inspired” ML models for dynamic systems.

The topics of interest for publication include but are not limited to:

  * Deep learning;
  * Gaussian processes;
  * Uncertainty quantification;
  * Data-driven techniques;
  * PDE solving;
  * Spatio-temporal forecasting;
  * Simulation;
  * Computational fluid dynamics;
  * Graphics;
  * Robotics.

*Manuscript Submission Information*

Manuscripts should be submitted online atwww.mdpi.com 
<https://www.mdpi.com/>byregistering 
<https://www.mdpi.com/user/register/>andlogging in to this website 
<https://www.mdpi.com/user/login/>. Once you are registered,click here 
to go to the submission form 
<https://susy.mdpi.com/user/manuscripts/upload/?journal=entropy>. 
Manuscripts can be submitted until the deadline. All submissions that 
pass pre-check are peer-reviewed. Accepted papers will be published 
continuously in the journal (as soon as accepted) and will be listed 
together on the special issue website. Research articles, review 
articles as well as short communications are invited. For planned 
papers, a title and short abstract (about 100 words) can be sent to the 
Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be 
under consideration for publication elsewhere (except conference 
proceedings papers). All manuscripts are thoroughly refereed through a 
single-blind peer-review process. A guide for authors and other relevant 
information for submission of manuscripts is available on 
theInstructions for Authors 
<https://www.mdpi.com/journal/entropy/instructions>page./Entropy/ 
<https://www.mdpi.com/journal/entropy/>is an international peer-reviewed 
open access monthly journal published by MDPI.

Please visit theInstructions for Authors 
<https://www.mdpi.com/journal/entropy/instructions>page before 
submitting a manuscript. TheArticle Processing Charge (APC) 
<https://www.mdpi.com/about/apc/>for publication in thisopen access 
<https://www.mdpi.com/about/openaccess/>journal is 1800 CHF (Swiss 
Francs). Submitted papers should be well formatted and use good English. 
Authors may use MDPI'sEnglish editing service 
<https://www.mdpi.com/authors/english>prior to publication or during 
author revisions.


    Keywords

  * deep learning
  * machine learning
  * PDE
  * neural networks
  * uncertainty quantification
  * physics-inspired meta-models
  * Gaussian processes

-- 
Prof. Patrick Gallinari
Sorbonne Universite - ISIR
4 place Jussieu, 75252 Paris Cedex 05, France
Tel: 33144277370

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/connectionists/attachments/20220520/863c7428/attachment.html>