Taking ROCKET on an Efficiency Mission: Multivariate Time Series Classification with LightWaveS

TL;DR

This paper introduces LightWaveS, a fast and scalable framework for multivariate time series classification that reduces feature usage from ROCKET while maintaining accuracy, enabling efficient deployment especially on edge devices.

Contribution

LightWaveS leverages wavelet scattering and feature selection to drastically reduce features and improve inference speed without sacrificing accuracy.

Findings

Achieves 9x to 53x inference speedup over ROCKET.

Uses only 2.5% of ROCKET features while maintaining accuracy.

Scales effectively across multiple compute nodes and input channels.

Abstract

Nowadays, with the rising number of sensors in sectors such as healthcare and industry, the problem of multivariate time series classification (MTSC) is getting increasingly relevant and is a prime target for machine and deep learning approaches. Their expanding adoption in real-world environments is causing a shift in focus from the pursuit of ever-higher prediction accuracy with complex models towards practical, deployable solutions that balance accuracy and parameters such as prediction speed. An MTSC model that has attracted attention recently is ROCKET, based on random convolutional kernels, both because of its very fast training process and its state-of-the-art accuracy. However, the large number of features it utilizes may be detrimental to inference time. Examining its theoretical background and limitations enables us to address potential drawbacks and present LightWaveS: a framework for accurate MTSC, which is fast both during training and inference. Specifically, utilizing wavelet scattering transformation and distributed feature selection, we manage to create a solution that employs just 2.5% of the ROCKET features, while achieving accuracy comparable to recent MTSC models. LightWaveS also scales well across multiple compute nodes and with the number of input channels during training. In addition, it can significantly reduce the input size and provide insight to an MTSC problem by keeping only the most useful channels. We present three versions of our algorithm and their results on distributed training time and scalability, accuracy, and inference speedup. We show that we achieve speedup ranging from 9x to 53x compared to ROCKET during inference on an edge device, on datasets with comparable accuracy.