Prediction of Time Series Data Based on Transformer with Soft Dynamic Time Wrapping

Kuo Hao Ho, Pei Shu Huang, I. Chen Wu, Feng Jian Wang

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

Abstract

It is a challenge to predict the long-term future data from time series data. This paper proposes to use a Transformer with soft dynamic time wrapping for early stopping criteria, called a soft-DTW Transformer. Our experiment in an open-source dataset HouseTwenty shows that the average prediction error rate with soft-DTW Transformer is 27.79%, greatly reduced from 45.70% for using SVR, a common time series method.

Original languageEnglish
Title of host publication2020 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728173993
DOIs
StatePublished - 28 Sep 2020
Event7th IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2020 - Taoyuan, Taiwan
Duration: 28 Sep 202030 Sep 2020

Publication series

Name2020 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2020

Conference

Conference7th IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2020
CountryTaiwan
CityTaoyuan
Period28/09/2030/09/20

Fingerprint Dive into the research topics of 'Prediction of Time Series Data Based on Transformer with Soft Dynamic Time Wrapping'. Together they form a unique fingerprint.

Cite this