Skip to main content
Cornell University
LASSP -  Laboratory of Atomic and Solid State Physics

Cornell Laboratory for Atomic and Solid State Physics

Eun-Ah Kim and Yi Zhang's Research Highlighted in Viewpoint

Viewpoint: Neural Networks Identify Topological Phases

A new machine-learning algorithm based on a neural network can tell a topological phase of matter from a conventional one.

A detailed characterization of phases of matter is at the forefront of research in condensed-matter and statistical physics. Although physicists have made incredible progress in the characterization of a wide variety of phases, the identification of novel topological phases remains challenging. Now, Yi Zhang and Eun-Ah Kim from Cornell University, New York [1], have taken a big-data approach to tackling this problem. In their work, thousands of microscopic “images” or “snapshots” of a phase, created using a special topography procedure, are fed into a machine-learning algorithm that is trained to decide whether these images come from a topological or a conventional phase of matter—exactly as modern computer vision algorithms are designed to tell cats from dogs in a picture.


Figure 1: Zhang and Kim’s machine-learning algorithm for identifying a topological phase of matter involves a procedure called quantum loop topography (QLT). The procedure builds a multidimensional image from several adjacent, triangular loops located at the pixels of snapshots of the phase’s electronic density (only one such snapshot is shown here). The QLT image is then fed into a neural network that is trained to determine whether the image corresponds to a topological phase or not.