2022
DOI: 10.1007/s42979-022-01346-z
|View full text |Cite
|
Sign up to set email alerts
|

Efficient Parabolic Optimisation Algorithm for Adaptive VQE Implementations

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
5
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
4
1

Relationship

0
5

Authors

Journals

citations
Cited by 5 publications
(5 citation statements)
references
References 26 publications
0
5
0
Order By: Relevance
“…At each iteration, the pool operators j scriptP are chosen from a complete minimal pool of size 2 n – 2, whose definition can be found in the Supporting Information. The question which selection criterion should be applied for picking the pool operators has been previously discussed, which includes considerations of utilizing the maximum gradient norm ,, or maximum impact on energy . While the latter approach was reported to show superior convergence properties, this work solely utilized the former one as in eq for the sake of comparability with the majority of previous studies.…”
Section: Methodsmentioning
confidence: 99%
See 2 more Smart Citations
“…At each iteration, the pool operators j scriptP are chosen from a complete minimal pool of size 2 n – 2, whose definition can be found in the Supporting Information. The question which selection criterion should be applied for picking the pool operators has been previously discussed, which includes considerations of utilizing the maximum gradient norm ,, or maximum impact on energy . While the latter approach was reported to show superior convergence properties, this work solely utilized the former one as in eq for the sake of comparability with the majority of previous studies.…”
Section: Methodsmentioning
confidence: 99%
“…The question which selection criterion should be applied for picking the pool operators has been previously discussed, which includes considerations of utilizing the maximum gradient norm ,, or maximum impact on energy . While the latter approach was reported to show superior convergence properties, this work solely utilized the former one as in eq for the sake of comparability with the majority of previous studies. The k th component of the energy gradient ∇ E can be rewritten as false( normal∇ E false) k = θ ψ M | normale i θ k normale i θ k | ψ M false| θ = 0 = 2 R e { ψ M | k | ψ M } …”
Section: Methodsmentioning
confidence: 99%
See 1 more Smart Citation
“…It is worth mentioning that the idea of parabola fitting has also been proposed in a recent work. 59 In each iteration, their method sets the optimization direction to the direction of the gradients and then employs parabola fitting to determine the minimum. Therefore, their method is gradient-based, while our method is gradient-free.…”
Section: Review and Motivationmentioning
confidence: 99%
“…where s i (k) is the score of unitary U i (θ i ) at the k-th iteration and θ i,init is the initial guess θ i . However, a large gradient does not necessarily indicate a large contribution to the total energy as pointed out by Armaos et al 47 Therefore, using a full optimization instead of gradients, it can be observed that in most cases where UCCSD operator pool is sufficient to obtain chemical accuracy, ES-VQE can generate similar or even more compact ansatzes compared with ADAPT-VQE. However, one typical drawback of ES-VQE is its one-shot scoring strategy.…”
Section: Qpe Implements the Time Evolution Operator =mentioning
confidence: 99%