Entropy-driven entanglement forging

Abstract

Simulating physical systems with variational quantum algorithms is a well-studied approach, but it is challenging to implement in current devices due to demands in qubit number and circuit depth. We show how limited knowledge of the system, namely the entropy of its subsystems, its entanglement structure or certain symmetries, can be used to reduce the cost of these algorithms with entanglement forging. To do so, we simulate a Fermi-Hubbard one-dimensional chain with a parametrized hopping term, as well as atomic nuclei ${}^{28}$Ne and ${}^{60}$Ti with the nuclear shell model. Using an adaptive variational quantum eigensolver we find significant reductions in both the maximum number of qubits (up to one fourth) and the amount of two-qubit gates (over an order of magnitude) required in the quantum circuits. Our findings indicate that our method, entropy-driven entanglement forging, can be used to adjust quantum simulations to the limitations of noisy intermediate-scale quantum devices.