Abstract

See the TCM graduate teaching page for further information.

In this series of graduate lectures we will study the application of deep neural networks to the approximation of wavefunctions. Since 2017 there has been a surge of interest in this area and this looks set to accelerate. Knowledge of quantum mechanics will be assumed up to early graduate level but familiarity with deep neural networks is not essential.

In the first half of this lecture we will motivate deep learning conceptually focussing on topics relevant to our end goal. This will not be a full introduction to deep neural networks (which would usually involve a practical element in any case) but should be sufficient to introduce the concepts we need. In the second half of the lecture we will discuss the application of deep learning to spin systems which came first historically.