FISES Archivos

Foro sobre Fisica Estadistica


Opciones: Vista Clásica

Use Monospaced Font
Mostrar las partes HTML
Mostrar todas las cabeceras de correo

Tema: [<< Primero] [< Prev] [Siguiente >] [Último >>]

Print Responder
Beatriz Seoane Bartolomé <[log in para visualizar]>
Tue, 8 Dec 2020 19:40:10 +0100
text/plain (1996 bytes) , text/html (5 kB)
---------- Forwarded message ---------
De: Ludovic Berthier <[log in para visualizar]>
Date: mar., 8 dic. 2020 19:16
Subject: [soft-matter] Machine Learning Glasses
To: <[log in para visualizar]>

Dear Colleagues,
Applications are invited to fill one postdoctoral fellowship on “Machine
Learning Glasses”, starting in 2021. The fellowship is for two years and is
funded by the Simons collaboration "Cracking The Glass Problem”.

The postdoc will be supervised by Ludovic Berthier (Montpellier, France and
Cambridge, UK) and Giulio Biroli (ENS Paris, France), and will develop
collaborations with Victor Bapst (DeepMind Google, London, UK).

The goal of the postdoc project is to combine two recent breakthroughs in
simulations of supercooled liquids and glasses (the swap Monte Carlo
algorithm and graph neural network developed at DeepMind and GoogleBrain)
to elucidate the physical properties of these systems at unprecedentedly
low-temperatures that are out of reach of conventional numerical approaches.

We are seeking candidates with good knowledge of statistical mechanics,
computer simulations of liquids, and/or machine learning techniques.

Further information about the postdoctoral position is available on the

We would be grateful if you would bring this announcement to the attention
of strong, creative individuals with a background in statistical physics
and/or machine learning and encourage them to apply.

The deadline for applications is January 15,  2021.

Ludovic Berthier & Giulio Biroli
For instructions on subscribing to and using
the Soft Matter Mailing List, please visit:

Please note: access to the mailman web interface is not possible from
outside the University of Leeds, but email commands usually work, see the
above link.