Heck reaction prediction using a transformer model based on a transfer learning strategy

Ling Wang; Chengyun Zhang; Renren Bai; Jianjun Li; Hongliang Duan

doi:10.1039/d0cc02657c

Heck reaction prediction using a transformer model based on a transfer learning strategy

Ling Wang, Chengyun Zhang, Renren Bai, Jianjun Li, Hongliang Duan

Research output: Contribution to journal › Article › peer-review

33 Citations (Scopus)

Abstract

A proof-of-concept methodology for addressing small amounts of chemical data using transfer learning is presented. We demonstrate this by applying transfer learning combined with the transformer model to small-dataset Heck reaction prediction. Introducing transfer learning significantly improved the accuracy of the transformer-transfer learning model (94.9%) over that of the transformer-baseline model (66.3%).

Original language	English
Pages (from-to)	9368-9371
Number of pages	4
Journal	Chemical Communications
Volume	56
Issue number	65
DOIs	https://doi.org/10.1039/d0cc02657c
Publication status	Published - 21 Aug 2020
Externally published	Yes

Access to Document

10.1039/d0cc02657c

Cite this

@article{d180433da2f94b8b88a54a558d49b869,

title = "Heck reaction prediction using a transformer model based on a transfer learning strategy",

abstract = "A proof-of-concept methodology for addressing small amounts of chemical data using transfer learning is presented. We demonstrate this by applying transfer learning combined with the transformer model to small-dataset Heck reaction prediction. Introducing transfer learning significantly improved the accuracy of the transformer-transfer learning model (94.9%) over that of the transformer-baseline model (66.3%).",

author = "Ling Wang and Chengyun Zhang and Renren Bai and Jianjun Li and Hongliang Duan",

note = "Publisher Copyright: {\textcopyright} The Royal Society of Chemistry.",

year = "2020",

month = aug,

day = "21",

doi = "10.1039/d0cc02657c",

language = "English",

volume = "56",

pages = "9368--9371",

journal = "Chemical Communications",

issn = "1359-7345",

publisher = "Royal Society of Chemistry",

number = "65",

}

TY - JOUR

T1 - Heck reaction prediction using a transformer model based on a transfer learning strategy

AU - Wang, Ling

AU - Zhang, Chengyun

AU - Bai, Renren

AU - Li, Jianjun

AU - Duan, Hongliang

N1 - Publisher Copyright: © The Royal Society of Chemistry.

PY - 2020/8/21

Y1 - 2020/8/21

N2 - A proof-of-concept methodology for addressing small amounts of chemical data using transfer learning is presented. We demonstrate this by applying transfer learning combined with the transformer model to small-dataset Heck reaction prediction. Introducing transfer learning significantly improved the accuracy of the transformer-transfer learning model (94.9%) over that of the transformer-baseline model (66.3%).

AB - A proof-of-concept methodology for addressing small amounts of chemical data using transfer learning is presented. We demonstrate this by applying transfer learning combined with the transformer model to small-dataset Heck reaction prediction. Introducing transfer learning significantly improved the accuracy of the transformer-transfer learning model (94.9%) over that of the transformer-baseline model (66.3%).

UR - http://www.scopus.com/inward/record.url?scp=85089614578&partnerID=8YFLogxK

U2 - 10.1039/d0cc02657c

DO - 10.1039/d0cc02657c

M3 - Article

C2 - 32672319

AN - SCOPUS:85089614578

SN - 1359-7345

VL - 56

SP - 9368

EP - 9371

JO - Chemical Communications

JF - Chemical Communications

IS - 65

ER -

Heck reaction prediction using a transformer model based on a transfer learning strategy

Abstract

Access to Document

Other files and links

Cite this