The importance of hyperparameters selection within small datasets

Parivash Ashrafi; Yi Sun; Neil Davey; Rod Adams; Marc B. Brown; Maria Prapopoulou; Gary Moss

doi:10.1109/IJCNN.2015.7280645

The importance of hyperparameters selection within small datasets

Parivash Ashrafi, Yi Sun, Neil Davey, Rod Adams, Marc B. Brown, Maria Prapopoulou, Gary Moss

School of Life and Health Sciences

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Gaussian Process is a Machine Learning technique that has been applied to the analysis of percutaneous absorption of chemicals through human skin. The normal, automatic method of setting the hyperparameters associated with Gaussian Processes may not be suitable for small datasets. In this paper we investigate whether a handcrafted search method of determining these hyperparameters is better for such datasets.

Original language	English
Title of host publication	2015 International Joint Conference on Neural Networks, IJCNN 2015
Publisher	Institute of Electrical and Electronics Engineers Inc.
Volume	2015-September
ISBN (Electronic)	9781479919604, 9781479919604, 9781479919604, 9781479919604
DOIs	https://doi.org/10.1109/IJCNN.2015.7280645
Publication status	Published - 28 Sept 2015
Event	International Joint Conference on Neural Networks, IJCNN 2015 - Killarney, Ireland Duration: 12 Jul 2015 → 17 Jul 2015

Other

Other	International Joint Conference on Neural Networks, IJCNN 2015
Country/Territory	Ireland
City	Killarney
Period	12/07/15 → 17/07/15

Keywords

covariance function hyperparameters
Gaussian Process
likelihood maximisation
QSAR

Access to Document

10.1109/IJCNN.2015.7280645

Cite this

@inproceedings{c62cf488ab5d41fcacc2adc0dee11680,

title = "The importance of hyperparameters selection within small datasets",

abstract = "Gaussian Process is a Machine Learning technique that has been applied to the analysis of percutaneous absorption of chemicals through human skin. The normal, automatic method of setting the hyperparameters associated with Gaussian Processes may not be suitable for small datasets. In this paper we investigate whether a handcrafted search method of determining these hyperparameters is better for such datasets.",

keywords = "covariance function hyperparameters, Gaussian Process, likelihood maximisation, QSAR",

author = "Parivash Ashrafi and Yi Sun and Neil Davey and Rod Adams and Brown, {Marc B.} and Maria Prapopoulou and Gary Moss",

year = "2015",

month = sep,

day = "28",

doi = "10.1109/IJCNN.2015.7280645",

language = "English",

volume = "2015-September",

booktitle = "2015 International Joint Conference on Neural Networks, IJCNN 2015",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

note = "International Joint Conference on Neural Networks, IJCNN 2015 ; Conference date: 12-07-2015 Through 17-07-2015",

}

Ashrafi, P, Sun, Y, Davey, N, Adams, R, Brown, MB, Prapopoulou, M & Moss, G 2015, The importance of hyperparameters selection within small datasets. in 2015 International Joint Conference on Neural Networks, IJCNN 2015. vol. 2015-September, 7280645, Institute of Electrical and Electronics Engineers Inc., International Joint Conference on Neural Networks, IJCNN 2015, Killarney, Ireland, 12/07/15. https://doi.org/10.1109/IJCNN.2015.7280645

The importance of hyperparameters selection within small datasets. / Ashrafi, Parivash; Sun, Yi; Davey, Neil et al.
2015 International Joint Conference on Neural Networks, IJCNN 2015. Vol. 2015-September Institute of Electrical and Electronics Engineers Inc., 2015. 7280645.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - The importance of hyperparameters selection within small datasets

AU - Ashrafi, Parivash

AU - Sun, Yi

AU - Davey, Neil

AU - Adams, Rod

AU - Brown, Marc B.

AU - Prapopoulou, Maria

AU - Moss, Gary

PY - 2015/9/28

Y1 - 2015/9/28

N2 - Gaussian Process is a Machine Learning technique that has been applied to the analysis of percutaneous absorption of chemicals through human skin. The normal, automatic method of setting the hyperparameters associated with Gaussian Processes may not be suitable for small datasets. In this paper we investigate whether a handcrafted search method of determining these hyperparameters is better for such datasets.

AB - Gaussian Process is a Machine Learning technique that has been applied to the analysis of percutaneous absorption of chemicals through human skin. The normal, automatic method of setting the hyperparameters associated with Gaussian Processes may not be suitable for small datasets. In this paper we investigate whether a handcrafted search method of determining these hyperparameters is better for such datasets.

KW - covariance function hyperparameters

KW - Gaussian Process

KW - likelihood maximisation

KW - QSAR

UR - http://www.scopus.com/inward/record.url?scp=84951075723&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.2015.7280645

DO - 10.1109/IJCNN.2015.7280645

M3 - Conference contribution

AN - SCOPUS:84951075723

VL - 2015-September

BT - 2015 International Joint Conference on Neural Networks, IJCNN 2015

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - International Joint Conference on Neural Networks, IJCNN 2015

Y2 - 12 July 2015 through 17 July 2015

ER -

The importance of hyperparameters selection within small datasets

Abstract

Other

Keywords

Access to Document

Other files and links

Fingerprint

Cite this