Transfer learning for time series classification using synthetic data generation

TL;DR

This paper introduces a novel transfer learning approach for time series classification that leverages a large synthetic dataset and regression tasks, achieving improved performance over traditional methods.

Contribution

It presents a new synthetic data generation method and demonstrates that using regression tasks as source tasks enhances transfer learning effectiveness for time series classification.

Findings

Synthetic dataset contains 15 million diverse time series.

Regression tasks outperform classification tasks as source tasks.

Method achieves superior accuracy compared to existing approaches.

Abstract

In this paper, we propose an innovative Transfer learning for Time series classification method. Instead of using an existing dataset from the UCR archive as the source dataset, we generated a 15,000,000 synthetic univariate time series dataset that was created using our unique synthetic time series generator algorithm which can generate data with diverse patterns and angles and different sequence lengths. Furthermore, instead of using classification tasks provided by the UCR archive as the source task as previous studies did,we used our own 55 regression tasks as the source tasks, which produced better results than selecting classification tasks from the UCR archive