InceptionTime: Finding AlexNet for Time Series Classification

TL;DR

This paper introduces InceptionTime, a deep learning model for time series classification that achieves accuracy comparable to the state-of-the-art HIVE-COTE but with significantly better scalability and training efficiency.

Contribution

InceptionTime is a novel ensemble of CNN models inspired by Inception-v4, offering a scalable alternative to HIVE-COTE for time series classification.

Findings

InceptionTime matches HIVE-COTE accuracy on benchmark datasets.

InceptionTime trains on millions of time series within hours.

InceptionTime significantly reduces training time compared to HIVE-COTE.

Abstract

This paper brings deep learning at the forefront of research into Time Series Classification (TSC). TSC is the area of machine learning tasked with the categorization (or labelling) of time series. The last few decades of work in this area have led to significant progress in the accuracy of classifiers, with the state of the art now represented by the HIVE-COTE algorithm. While extremely accurate, HIVE-COTE cannot be applied to many real-world datasets because of its high training time complexity in O(N2 * T4) for a dataset with N time series of length T. For example, it takes HIVE-COTE more than 8 days to learn from a small dataset with N = 1500 time series of short length T = 46. Meanwhile deep learning has received enormous attention because of its high accuracy and scalability. Recent approaches to deep learning for TSC have been scalable, but less accurate than HIVE-COTE. We introduce InceptionTime - an ensemble of deep Convolutional Neural Network (CNN) models, inspired by the Inception-v4 architecture. Our experiments show that InceptionTime is on par with HIVE-COTE in terms of accuracy while being much more scalable: not only can it learn from 1,500 time series in one hour but it can also learn from 8M time series in 13 hours, a quantity of data that is fully out of reach of HIVE-COTE.