Accelerated electronic structure with deep learning

Our group (CLEAN@NRC) seeks a postdoctoral fellow for a project linking deep neural networks with electronic structure theory. So far, we have shown that deep networks can be used to solve the Schrodinger Equation (, classical spin models (, and 2d-materials such as graphene and boron-nitride (

The project will explore the use of our recently reported extensive deep neural networks ( to the electronic structure problem within the density functional theory. The objective is to show that EDNN can outperform "traditional" electronic structure methods by a factor of 1,000,000. We will generate predictive results 1000 times faster than is currently possible and work on problems 1000 times larger than can currently be modelled.

Salary = $65,000 CAD / year for two years + relocation allowance + conference travel allowance

Location = Ottawa, Ontario, Canada

Coffee = free


Applicants should provide a CV, Statement of Interest, and contact information for 3 references

Application deadline = 10 May 2018