Puede ser de interés para alguien de la lista.

A postdoctoral position (for one year renewable up to 3 years) will be available in our group at the Universidad Complutense of Madrid (Dept. of Theoretical Physics) to work at the frontier between statistical mechanics and machine learning.

The candidate will work on both computational and theoretical aspects of generative models that are of interest for both statistical physicists and computer scientists. The project aims to characterize both equilibrium and non-equilibrium aspects of the Restricted Boltzmann Machine, the eventual learning dynamics and to improve the learning process. The successful candidate will have to develop numerical methods through GPU libraries on python and to investigate analytically the learning behavior and the phase diagram of the RBM.

The candidate must have a PhD in physics, mathematics or computer science and a strong experience amongst one of the following fields: statistical mechanics, machine learning, spin glasses. Programming, mathematical modeling and data-analysis skills are required.

Collaboration and integration within the research group will be encouraged.

Applications are welcome until the position is filled, but priority will be given for applications sent before the 20th of May 2021. This position is expected to start around the last semester of 2021 (beginning of September at the soonest).

Potential candidates should send a CV, a letter presenting his motivation and contact for two possible reference letters.

--

Beatriz Seoane Bartolomé

Departamento de Física Teórica,

Universidad Complutense de Madrid,

---------- Forwarded message ---------

De:**Aurélien Decelle** <[log in para visualizar]>

Date: lun, 26 abr 2021 a las 14:45

Subject: PostDoc position in statistical physics

To:

De:

Date: lun, 26 abr 2021 a las 14:45

Subject: PostDoc position in statistical physics

To:

Dear colleagues,

The candidate will work on both computational and theoretical aspects of generative models that are of interest for both statistical physicists and computer scientists. The project aims to characterize both equilibrium and non-equilibrium aspects of the Restricted Boltzmann Machine, the eventual learning dynamics and to improve the learning process. The successful candidate will have to develop numerical methods through GPU libraries on python and to investigate analytically the learning behavior and the phase diagram of the RBM.

The candidate must have a PhD in physics, mathematics or computer science and a strong experience amongst one of the following fields: statistical mechanics, machine learning, spin glasses. Programming, mathematical modeling and data-analysis skills are required.

Collaboration and integration within the research group will be encouraged.

Applications are welcome until the position is filled, but priority will be given for applications sent before the 20th of May 2021. This position is expected to start around the last semester of 2021 (beginning of September at the soonest).

Potential candidates should send a CV, a letter presenting his motivation and contact for two possible reference letters.

For further information or submissions, please contact Aurélien Decelle, [log in para visualizar].

Sincerely,

Aurélien Decelle.

--

Beatriz Seoane Bartolomé

Departamento de Física Teórica,

Universidad Complutense de Madrid,

Plaza de Ciencias, 1,

28040 Madrid,

Spain

Spain