2020
DOI: 10.1103/physrevresearch.2.032060
|View full text |Cite
|
Sign up to set email alerts
|

Wave-function positivization via automatic differentiation

Abstract: We introduce a procedure to systematically search for a local unitary transformation that maps a wave function with a nontrivial sign structure into a positive-real form. The transformation is parametrized as a quantum circuit compiled into a set of one-and two-qubit gates. We design a cost function that maximizes the average sign of the output state and removes its complex phases. The optimization of the gates is performed through automatic differentiation algorithms, widely used in the machine learning commu… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

2
18
0

Year Published

2020
2020
2024
2024

Publication Types

Select...
5
3
1

Relationship

0
9

Authors

Journals

citations
Cited by 32 publications
(20 citation statements)
references
References 23 publications
(28 reference statements)
2
18
0
Order By: Relevance
“…It may therefore be possible to obtain fine-tuned models and QMC methods that lead to a small enough to be practically useful. More generally, it may be possible to search for such models and methods algorithmically, thus easing the intrinsic sign problem [10,[122][123][124]. We also note that the results presented in this paper do not exclude approaches to the sign problem based on a modified or constrained Monte Carlo sampling [13,125,126], as well as machine-learning-aided QMC [127], and infinite-volume diagrammatic QMC [128].…”
Section: Discussion and Outlookmentioning
confidence: 93%
“…It may therefore be possible to obtain fine-tuned models and QMC methods that lead to a small enough to be practically useful. More generally, it may be possible to search for such models and methods algorithmically, thus easing the intrinsic sign problem [10,[122][123][124]. We also note that the results presented in this paper do not exclude approaches to the sign problem based on a modified or constrained Monte Carlo sampling [13,125,126], as well as machine-learning-aided QMC [127], and infinite-volume diagrammatic QMC [128].…”
Section: Discussion and Outlookmentioning
confidence: 93%
“…Other methods for generating positive-valued decompositions of the partition function include, e.g., resummation techniques wherein negativevalued weights in the decomposition are grouped together with positive ones to form positive ``superweights"" that can in turn be treated as probabilities in a quantum Monte Carlo algorithm [35,13,19]. Other methods also include applying a constantdepth quantum circuit [32]. These other methods are beyond the scope of this paper.…”
mentioning
confidence: 99%
“…We assume that these low-energy states are similar to the ground state of the un-truncated Hamiltonian in regions far from the boundary. 5 Hence, the low-energy states correspond to excitations localized near the boundary or degenerate ground states.…”
Section: Anomalous Symmetry Action At a Boundarymentioning
confidence: 99%
“…While the magic in a many-body state and the notion of a sign problem are promising metrics for the quantum complexity of states, they are notoriously challenging to study analytically and numerically, although substantial progress has been made [1,[4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20]. We therefore propose a simplification by imposing symmetry constraints.…”
Section: Introductionmentioning
confidence: 99%