A subpolynomial-time algorithm for the free energy of one-dimensional quantum systems in the thermodynamic limit - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

A subpolynomial-time algorithm for the free energy of one-dimensional quantum systems in the thermodynamic limit

Résumé

We introduce a classical algorithm to approximate the free energy of local, translation-invariant, one-dimensional quantum systems in the thermodynamic limit of infinite chain size. While the ground state problem (i.e., the free energy at temperature $T = 0$) for these systems is expected to be computationally hard even for quantum computers, our algorithm runs for any fixed temperature $T > 0$ in subpolynomial time, i.e., in time $O((\frac{1}{\varepsilon})^{c})$ for any constant $c > 0$ where $\varepsilon$ is the additive approximation error. Previously, the best known algorithm had a runtime that is polynomial in $\frac{1}{\varepsilon}$ where the degree of the polynomial is exponential in the inverse temperature $1/T$. Our algorithm is also particularly simple as it reduces to the computation of the spectral radius of a linear map. This linear map has an interpretation as a noncommutative transfer matrix and has been studied previously to prove results on the analyticity of the free energy and the decay of correlations. We also show that the corresponding eigenvector of this map gives an approximation of the marginal of the Gibbs state and thereby allows for the computation of various thermodynamic properties of the quantum system.
Fichier principal
Vignette du fichier
2209.14989.pdf (398.06 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03852131 , version 1 (14-11-2022)

Licence

Paternité

Identifiants

Citer

Hamza Fawzi, Omar Fawzi, Samuel O. Scalet. A subpolynomial-time algorithm for the free energy of one-dimensional quantum systems in the thermodynamic limit. ITCS 2023 - Innovations in Theoretical Computer Science, Jan 2023, Boston, United States. pp.1-33. ⟨hal-03852131⟩
41 Consultations
17 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More