Feature Selection for Multivariate Time Series via Network Pruning

TL;DR

This paper introduces Neural Feature Selector (NFS), a novel end-to-end neural network component for feature selection in multivariate time series data, improving interpretability and robustness while maintaining competitive predictive performance.

Contribution

The paper proposes NFS, a new neural network module that performs feature selection in MTS data using a decomposed convolution approach, addressing high dimensionality challenges.

Findings

NFS achieves comparable accuracy to state-of-the-art methods.

NFS effectively selects relevant features, enhancing interpretability.

NFS demonstrates robustness over autoencoder-based feature selection methods.

Abstract

In recent years, there has been an ever increasing amount of multivariate time series (MTS) data in various domains, typically generated by a large family of sensors such as wearable devices. This has led to the development of novel learning methods on MTS data, with deep learning models dominating the most recent advancements. Prior literature has primarily focused on designing new network architectures for modeling temporal dependencies within MTS. However, a less studied challenge is associated with high dimensionality of MTS data. In this paper, we propose a novel neural component, namely Neural Feature Selector (NFS), as an end-2-end solution for feature selection in MTS data. Specifically, NFS is based on decomposed convolution design and includes two modules: firstly each feature stream (a stream corresponds to an univariate series of MTS) within MTS is processed by a temporal CNN independently; then an aggregating CNN combines the processed streams to produce input for other downstream networks. We evaluated the proposed NFS model on four real-world MTS datasets and found that it achieves comparable results with state-of-the-art methods while providing the benefit of feature selection. Our paper also highlights the robustness and effectiveness of feature selection with NFS compared to using recent autoencoder-based methods.