A Quantum Phase Estimation Toolbox

The quantum phase estimation (QPE) algorithm is the most studied algorithm for evaluating ground state energies of molecules with fault tolerant (FT) quantum computers. Recently, significant efforts have reduced the resource costs of QPE for the FT quantum hardware. We provide an open-source python package, the QPE Toolbox, which combines quantum chemistry and tensor networks (TNs) methods to produce and evaluate the quantum circuits of the QPE pipeline.

We articulate the toolbox on top of the TN library Quimb to provide: (1) classical preprocessing strategies focused on initializing and setting up QPE circuits, (2) an internal circuit-level simulator, and (3) a list of postprocessing functionalities for retrieving final energies.

For the preprocessing stage, the QPE Toolbox offers fermionic encodings based on OpenFermion and PySCF. From the qubit Hamiltonian in TN form, we find a candidate state with high overlap to the ground state through DMRG, and derive the associated evolution circuit either from Trotterization or block encoding through Linear Combination of Unitaries to produce a Walk Operator. Finally, the QPE algorithm is assembled using this circuit to achieve different versions of QPE and Single-Ancilla Robust Phase Estimation.

We believe that the QPE Toolbox will be a valuable open-source package for the quantum computing and chemistry communities, preparing the stage for fault-tolerant quantum chemistry calculations.