Quantum simulation with atom arrays - from lattice gauge theories to complexity of a many-body system
Date: 11 JANUARY 2024 from 14:00 to 15:00
Event location: IR-2A
Arrays of neutral atoms laser-coupled to highly excited (Rydberg) states are presently a very promising platform in the realm of the so-called programmable quantum systems - quantum computers and quantum simulators. In these devices, interactions between microscopic components can be controlled at the single quantum level, uncovering unprecedented opportunities to investigate the quantum many-body problem both within the context of traditional statistical physics, as well as from a information theoretic viewpoint.
After a non-technical introduction to the overall state of the art, I will discuss two of our research lines in the field.
The first concerns the quantum simulation of lattice gauge theories. Such theories, originally devised by Wilson as possible regularizations of continuum ones, have played a pivotal role in several fields, including particle (Lattice QCD) and condensed matter physics (spin liquids). However, they often present insurmountable challenges from a computational standpoint, especially in the context of real-time dynamics. I will show how it is possible to realize the latter in quantum simulators, challenging what we can compute with state of the art computational techniques. These new theoretical ideas are also a fertile ground to develop an understanding of gauge theory phenomena utilizing entanglement, as I will illustrate in the context of the Schwinger model. I will conclude this part with a critical assessment of what can and cannot be done in terms of mid term perspectives.
The second research line concerns the probing capabilities offered by such systems. While it is often possible to gather large amount of data from quantum simulators, extracting information in an assumption-free manner (like, e.g., what is routinely done in other branches of science involving big data) is very underdeveloped. I will present a toolbox to carry this task out, combining the mathematical framework of network theory, with non-parametric learning methods. These methods allow for statistically accurate ways of validating quantum simulators and computers, and estimate their data complexity, as I will show in the context of both theory and experimental data.
First part: 1902.09551, 2205.13000
Second part: 2301.13216