Speaker
Description
In this talk, I will discuss the difference between a classical reservoir (which adds no additional quantum degrees of freedom) and a quantum reservoir (which does add quantum degrees of freedom). I will then explain why the classical reservoir is preferred for pure-state creation, because it cannot create mixed states. Then I will describe an efficient algorithm for ground-state preparation of the Hubbard model that avoids the need for Jordan-Wigner strings in the ansatz and just employs hopping terms and on-site correlations in preparing the ground state. I believe this is perhaps the most efficient way to create ground states. I will also show how this approach compares to other methods for ground-state preparation such as couple-cluster approaches. I may even have some preliminary results for how well this methodology works in quantum chemistry ground-state preparation. The approach does require a global optimization strategy, which is complicated to implement, but this is ameliorated by the fact that the number of parameters can be kept rather small (empirically appearing to grow linearly with the number of sites/orbitals in the system). Our results have recently appeared in Phys. Rev. B 111, 235152 (20025).
This work was completed in collaboration with Zach He, Lex Kemper, Lorenzo Del Re, and Dominika Zgid.
