Deep Physics MSc Projects — Skoltech MSc class of 2020 

We still have two openings to advise MSc students.  

Deep learning and quantum computing. We have several ongoing and new projects to select from that apply deep learning to understand physics.  We also use statistical mechanics and quantum theory as a mathematical tool to understand e.g. the informatic properties of deep neural networks.  For example, quantum computers provide new classes of neural networks that appear to be exponentially difficult to simulate using classical computers; on going work is geared towards refining quantum deep learning algorithms.  

  1. Applications of deep learning to predict properties of quantum systems
  2. Enhancement of deep learning algorithms using real-world quantum computers and programable quantum simulators
  3. Applications of deep learning to understand properties of statistical physics, particularly in the setting of complex networks (see below)

MSc students at Skoltech who are interested in (i) writing computer programs, or (ii) in doing theory which is driven by numerical experiments or (iii) both should get into contact. These projects are particularly suited to students who wish to publish a research paper in a leading journal.  

MSc project in “Deep Learning in Complex Networks”

We are looking for a first year MSc student who is interested in machine learning to assist our ongoing research projects with Saeed Osat, Jacob Biamonte and others in Deep Quantum Labs.  The prospective student should be motivated to assist in publications in leading journals.

A tantamount discovery occurred late in the last century.  A wide range of complex systems where shown to be modelled using a theory of graphs augmented with statistical mechanics.  This approach unveiled striking commonalities between a priori distant fields, illustrating commonality in scaling behaviour between various biological, social and technological networks.  Although techniques in machine learning have been successfully applied in the theory of complex networks, applications of deep learning in this rapidly emerging domain of science is in its nascent days.

Our ongoing work is aimed at augmenting the current complex networks tool set with techniques and methods found in deep learning.  A prospective student should be motivated to assist in the publication of journal papers.  Students will be given hands on tutorials on the relevant deep learning and complex network theory needed to for their work.

More details about the research and the people in the lab can be found at DeepQuantum.AI

Leave a Reply

Your email address will not be published. Required fields are marked *