Adiabatic Quantum Computation with the Fermionic Position Space Schr\"odinger Equation

TL;DR

This paper presents a novel encoding method for simulating the fermionic Schrödinger equation in a finite lattice using adiabatic quantum computation, addressing the challenge of implementing the kinetic energy operator with local potentials.

Contribution

It introduces operator filtering and entanglement gadgets to efficiently encode fermionic operators and implements a Laplacian operator with polynomial complexity in a finite volume setting.

Findings

Encoding scales as O(An 2^D) for the Laplacian

Finite volume context maintains a spectral gap

Polynomial time complexity with box size

Abstract

The efficient encoding of the fermionic Schr\"odinger equation as a spin system Hamiltonian is a long-term problem. I describe an encoding for the fermionic position space Schr\"odinger equation on a finite-volume periodic lattice with a local potential. The challenging part of the construction is the implementation of the kinetic energy operator, which is essentially the Laplacian. The finite difference implementation on the lattice combines contributions from neighboring lattice sites, which is complicated by fermionic exchange symmetry. Two independently useful techniques developed here are operator filtering and entanglement gadgets. Operator filtering is useful when a simple operator acting on a subspace of the full Hilbert space has a desired set of interactions. Occupation suppression of the complement of the subspace then filters away unwanted contributions of the operator. Entanglement gadgets encode the same information differently in two sets of qubits. We may then independently choose the most efficient encoding for operators acting on the qubits. The construction for the Laplacian described here has $\mathcal{O}\left(An 2^D\right)$ cost in bounded Pauli weight terms where $A$ is the number of identical spinless fermions, $N=2^n$ is the number of lattice points in each direction, and $D$ is the number of dimensions. The finite volume context protects the gap between the ground state and the first excited state, yielding polynomial time complexity with the box size.