Lee-Yang Tensors in Quantum Information and Algorithms for Local Hamiltonian Problems

Loading...
Thumbnail Image

Journal Title

Journal ISSN

Volume Title

Publisher

University of Waterloo

Abstract

Local Hamiltonian problems are central to both quantum complexity theory and quantum many-body physics: ground state problems are the natural quantum constraint satisfaction problems, and computing the partition function reveals thermal equilibrium properties. A turning point in the theory of local Hamiltonians was Lee and Yang’s original study relating roots of the partition function to phase transitions [LY52]. In this thesis, we investigate the role of complex tensors which satisfy a Lee-Yang condition in quantum information theory. We find that the class of Lee-Yang tensors is closed under natural operations such as certain post-selective measurements, tensor products, tensor contraction, and the application of certain single-qubit quantum channels. In analogy with the Perron-Frobenius theorem, we show that Hermitian operators with the Lee-Yang property have simple extremal eigenvalues. We show that there are quasipolynomial-size quantum circuits for preparing Lee-Yang tensors. Furthermore, we present new algorithms for specific families of quantum many-body systems. The EPR Hamiltonian is a model in which local terms energetically favor EPR states |00⟩+|11⟩and which includes quantum Max Cut on bipartite graphs as a special case. We provide numerical evidence for an efficient adiabatic quantum algorithm for preparing the ground state of the EPR Hamiltonian. For the ferromagnetic XY model, we give an efficient quantum algorithm for preparing the Gibbs state, thereby extending a classical algorithm due to Bravyi and Gosset [BG17]. We give an efficient classical algorithm for simulating stoquastic systems when the interaction graph has a sufficiently rich family of automorphisms. For the classical Ising model, we establish that a polynomial built from powers of the Hamiltonian satisfies a zero-freeness condition.

Description

Citation

Endorsement

Review

Supplemented By

Referenced By