Abstract
One of the most important challenges in condensed matter physics is to understand how superconductivity is stabilized under realistic microscopic interactions. I will discuss some progress we've made on understanding pairing in the paradigmatic square lattice Fermi-Hubbard model. First I will describe Hidden fermion pfaffian states (HFPS), a neural network wave function which can be used to model paired states, and how we improved the scaling complexity for large systems. Then I will discuss how we have used HFPS to learn about the Fermi-Hubbard model. Through describing our results, I will highlight some of the advantage of NQS over other high-precision numerical methods, and discuss some challenges that NQS may be able to address.
