Efficient support vector regression with reduced training data

Ling Cen, Quang Hieu Vu, Dymitr Ruta

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

5 Scopus citations

Abstract

Support Vector Regression (SVR) as a supervised machine learning algorithm have gained popularity in various fields. However, the quadratic complexity of the SVR in the number of training examples prevents it from many practical applications with large training datasets. This paper aims to explore efficient ways that maximize prediction accuracy of the SVR at the minimum number of training examples. For this purpose, a clustered greedy strategy and a Genetic Algorithm (GA) based approach are proposed for optimal subset selection. The performance of the developed methods has been illustrated in the context of Clash Royale Challenge 2019, concerned with decks' win rate prediction. The training dataset with 100,000 examples were reduced to hundreds, which were fed to SVR training to maximize model prediction performance measured in validation R2 score. Our approach achieved the second highest score among over hundred participating teams in this challenge.

Original languageBritish English
Title of host publicationProceedings of the 2019 Federated Conference on Computer Science and Information Systems, FedCSIS 2019
EditorsMaria Ganzha, Leszek Maciaszek, Leszek Maciaszek, Marcin Paprzycki
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages15-18
Number of pages4
ISBN (Electronic)9788395541605
DOIs
StatePublished - Sep 2019
Event2019 Federated Conference on Computer Science and Information Systems, FedCSIS 2019 - Leipzig, Germany
Duration: 1 Sep 20194 Sep 2019

Publication series

NameProceedings of the 2019 Federated Conference on Computer Science and Information Systems, FedCSIS 2019

Conference

Conference2019 Federated Conference on Computer Science and Information Systems, FedCSIS 2019
Country/TerritoryGermany
CityLeipzig
Period1/09/194/09/19

Keywords

  • Clash Royale
  • Greedy search
  • K-means clustering
  • R-squared metric
  • Support Vector Regression (SVR)

Fingerprint

Dive into the research topics of 'Efficient support vector regression with reduced training data'. Together they form a unique fingerprint.

Cite this