On model reduction for quantum dynamics: symmetries and invariant subspaces

Abstract

Simulation of quantum dynamics is a grand challenge of computational physics. In this work we investigate methods for reducing the demands of such simulation by identifying reduced-order models for dynamics generated by parameterized quantum Hamiltonians. In particular, we first formulate an algebraic condition that certifies the existence of invariant subspaces for a model defined by a parameterized Hamiltonian and an initial state. Following this we develop and analyze two methods to explicitly construct a reduced-order model, if one exists. In addition to general results characterizing invariant subspaces of arbitrary finite dimensional Hamiltonians, by exploiting properties of the generalized Pauli group we develop practical tools to speed up simulation of dynamics generated by certain spin Hamiltonians. To illustrate the methods developed we apply them to several paradigmatic spin models.

26 pages, 2 figures, 3 algorithms. Version 2 is published version and contains new results on stability of invariant subspaces to perturbations

References in corpus (5)

• Matrix Product States, Projected Entangled Pair States, and variational renormalization group methods for quantum spin systems
• Real time evolution using the density matrix renormalization group
• Quantum simulation of time-dependent Hamiltonians and the convenient illusion of Hilbert space
• Quantum filter reduction for measurement-feedback control via unsupervised manifold learning
• Derivation of Maxwell-Bloch-type equations by projection of quantum models