Postdoc in Machine Learning for Classical Simulation of Many-Body Quantum Systems and Quantum Computing

The Chair for Artificial Intelligence and Quantum Physics at the Theoretical Physics Laboratory (CPHT) of Ecole Polytechnique, Paris, is opening a call for 1 Postdoctoral fellow at the interface between Machine Learning, Classical Simulation of Many-Body Quantum Systems and Quantum Computing.

The chair is led by Prof. Filippo Vicentini, and currently groups 3 other PhD students, 1 Postdoc and some interns. Filippo Vicentini is recognized as one of the most prominent figures in the nascent field of Neural-Network quantum states, having contributed to the development of several fundamental ideas to describe systems out-of-equilibrium, as well as several algorithms for simulating their dynamics. F.V. also leads the NetKet open-source collaboration, developing the most widely used software in the field. The group is supported by CNRS, has been awarded an ANR Junior Researcher grant and a JRC startup fund. The group is also affiliated to INRIA Saclay and College de France and regularly visits and partecipates in joint activities with those institutions. The group maintains strong ties with the groups of Prof. Giuseppe Carleo and Vincenzo Savona at EPFL, and other research groups overseas.

The objective of the group is to expand the boundaries of what can be simulated with Classical and Quantum algorithms, leveraging ideas from optimisation theory and Machine Learning, and to apply such algorithms to answer open problems in fundamental quantum science (Entanglement Phase Transitions, Transport problems...) as well as in condensed matter physics (correlated electrons in solid state and chemistry). Core interests of the group at this stage are (i) methodological improvements to simulate the quantum dynamics and/or accessing the spectra of correlated systems, (ii) methodological improvements to the simulation of systems out of equilibrium, (iii) application of said methods to correlated electrons, (iv) foundational understanding of neural quantum states and their resource theory/representability arguments and (v) connections to quantum computing and hybrid classical-quantum algorithms.

Candidates are expected to have (or be awaiting its award) in the area of Theoretical/Computational Physics and a strong expertise in numerical methods for one of correlated electrons, systems driven far from equilibrium or closely related problems. Experience with Machine Learning is preferred. Candidates must show a clear, genuine interest and plausible research plan to combine their interests with the research of the group. Experience with Influence Functional, Variational Monte Carlo or Quantum Algorithms will also be evaluated positively. Experience supervising interns/younger students will also be evaluated positively.

Succesful candidates will be expected to work on some of the topics mentioned above, but will also be encouraged and supported to combine their previous experience in this new setting. An ideal candidate would also express an interest in taking on co-supervising master interns and collaborating with PhD students. The group is strongly engaged in the open-source philosophy, and postdocs are expected to take on a role curating and contributing algorithms to the larger scientific community.

Starting date: as soon as agreed. International and female applicants are encouraged to apply. The contract is for a 2+1 years position. Candidates will be reviewed on a rolling basis and the position will remain open until filled.

To get an idea of the research direction of group, you can look at recent papers Google Scholar. For informations about the application, send an email to filippo.vicentini [at] polytechnique [dot] edu