Quantum k-fold cross-validation for nearest neighbor classification algorithm

Cross-validation is one of the important tools in machine learning, which is generally used for performance evaluation. It uses different portions of the data to test and train a model on different iterations, which leads to a high computational cost. In this paper, we present a quantum version of k...

Celý popis

Uložené v:

Podrobná bibliografia
Vydané v:	Physica A Ročník 611; s. 128435
Hlavní autori:	Li, Jing, Gao, Fei, Lin, Song, Guo, Mingchao, Li, Yongmei, Liu, Hailing, Qin, Sujuan, Wen, QiaoYan
Médium:	Journal Article
Jazyk:	English
Vydavateľské údaje:	Elsevier B.V 01.02.2023
Predmet:	Cross-validation Quantum computing Quantum machine learning Quantum nearest neighbor algorithm Quantum computing Quantum machine learning Cross-validation Quantum nearest neighbor algorithm
ISSN:	0378-4371, 1873-2119
On-line prístup:	Získať plný text
Tagy:	Pridať tag Žiadne tagy, Buďte prvý, kto otaguje tento záznam!

Popis
Shrnutí:	Cross-validation is one of the important tools in machine learning, which is generally used for performance evaluation. It uses different portions of the data to test and train a model on different iterations, which leads to a high computational cost. In this paper, we present a quantum version of k-fold cross-validation to choose a good parameter for the nearest neighbor classification algorithm with a threshold t, where the classification performance is estimated efficiently. With the help of amplitude amplification and estimation, the proposed quantum algorithm achieves a polynomial speedup on the number of samples over its classical counterpart. •A quantum k-fold cross-validation is proposed to choose a good parameter for the NN classifier efficiently.•The proposed quantum algorithm achieves a polynomial speedup over its classical counterpart.•A modified quantum NN classifier with a threshold t is presented.
ISSN:	0378-4371 1873-2119
DOI:	10.1016/j.physa.2022.128435