The 10th school on mathematical physics will take place from the 27th to the 31st of May 2019. It will introduce modern machine learning (ML) techniques for studying classical and quantum many-body problems encountered in condensed matter, quantum information and technology, and related fields. The lectures will emphasize relations between statistical physics, quantum many-body physics, and machine learning, while tutorials will include hands-on experience in programming with applications.

Topics to be covered include lattice models for statistical physics, Monte Carlo methods, supervised and unsupervised learning, neural networks, energy-based generative models, modern unsupervised learning models, reinforcement learning, and deep learning. Via tutorials, students will be exposed to important computational tools such as Python, Pytorch and/or TensorFlow machine learning libraries, as well as modern machine learning software tailored specifically to address problems in quantum physics such as QuCumber and NetKet.

The school is aimed at undergraduate, graduate and PhD students and researchers interested in the topic. Attendees are expected to have basic knowledge of linear algebra, calculus, probability and statistics, statistical mechanics, and quantum mechanics. Some coding experience in any programming language is desired but not required.

The school will consist of a series of morning lectures given by the invited speakers. The participants will have the possibility to give talks or present posters in the afternoon sessions. We are also studying the possibility to have problem sessions in the afternoons. The morning lectures will be held in English, afternoon lectures can be held in English or Spanish.

• Juan Carrasquilla (Vector Institute, University of Waterloo, Canada),
• Roger Melko (University of Waterloo and Perimeter Institute, Canada),
• Giacomo Torlai (Flatiron Institute, Simons Foundation, USA).

Titles of the mini courses:

• Statistical mechanics and Monte Carlo simulations
• Introduction to elements of machine learning
• Generative models, ML-inspired representations of quantum state, and applications in quantum physics and technology

Videos:

Videos of the talks are available at the youtube channel of the Department of Physics of the Universidad de los Andes.

Suggested literature:

• A high-bias, low-variance introduction to Machine Learning for physicists
• Deep learning book
• Neural networks and deep learning
• The Elements of Statistical Learning
• Monte Carlo Methods in Statistical Physics

If you are interested in participating in the school, please register before the 19th of May 2019 here: el formulario de inscripción.

If you need further information, let us know by sending an e-mail to the organizers . If you are interested in giving a talk, please contact the organizers before the 14th of April 2019.

There will be no registration fee. Limited financial support for some participants from outside of Bogotá might be available. Participants interested in applying for financial support should indicate this clearly in the comments section in the registration form before the 31st of March 2019.

Organizers:

Acknowledgement: We want to express our gratitude for financial support by

• Department of Mathematics of the Universidad de Los Andes
• Department of Physics of the Universidad de Los Andes