2026 Chhabra-Landau Lecture

Combining Monte Carlo and Tensor Networks to beat the Quantum Exponential

Steven White, Distinguished Prof. of Physics and Astronomy, University of California, Irvine, Member of the National Academy of Sciences, will deliver the 2026 Chhabra-Landau lecture in the Franklin College of Arts and Sciences department of physics and astronomy Center for Simulational Physics.

The key difficulty in understanding quantum systems is the exponentially large space of states when

there are many particles. Monte Carlo methods have been used for most of a century to tackle

exponentially large spaces, but for most quantum systems they encounter the "sign problem",

limiting their usefulness. Tensor networks (including the Density Matrix Renormalization Group,

DMRG) use a completely different idea to deal with the exponential: compression of the wave

function, which collapses the exponential to a number of parameters that only grows linearly with

the system size. The two key limitations of tensor networks are high entanglement--they are low

entanglement methods--and dimensionality--the most efficient and accurate algorithms are tied to

1D problems. Here I will give elementary introductions to both Monte Carlo and Tensor Networks

and then describe two different problems where we combined the power of these approaches to

make substantial progress. In the first project, we utilized both Monte Carlo and DMRG as

complementary tools to study high temperature superconductivity in the Hubbard model. The

second application combines Monte Carlo and tensor networks in the same hybrid algorithm, in

order to study finite temperature quantum systems.

Via Zoom: https://zoom.us/j/99879004873?pwd=Vkp2dHJDdU9tcnpNUWp5SFk4QVIvQT09