Transparent Networks for Multivariate Time Series

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kim, Minkyu, Lee, Suan, Kim, Jinho
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915676884566016
author Kim, Minkyu
Lee, Suan
Kim, Jinho
author_facet Kim, Minkyu
Lee, Suan
Kim, Jinho
contents Transparent models, which provide inherently interpretable predictions, are receiving significant attention in high-stakes domains. However, despite much real-world data being collected as time series, there is a lack of studies on transparent time series models. To address this gap, we propose a novel transparent neural network model for time series called Generalized Additive Time Series Model (GATSM). GATSM consists of two parts: 1) independent feature networks to learn feature representations, and 2) a transparent temporal module to learn temporal patterns across different time steps using the feature representations. This structure allows GATSM to effectively capture temporal patterns and handle varying-length time series while preserving transparency. Empirical experiments show that GATSM significantly outperforms existing generalized additive models and achieves comparable performance to black-box time series models, such as recurrent neural networks and Transformer. In addition, we demonstrate that GATSM finds interesting patterns in time series.
format Preprint
id arxiv_https___arxiv_org_abs_2410_10535
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Transparent Networks for Multivariate Time Series
Kim, Minkyu
Lee, Suan
Kim, Jinho
Machine Learning
Computers and Society
Transparent models, which provide inherently interpretable predictions, are receiving significant attention in high-stakes domains. However, despite much real-world data being collected as time series, there is a lack of studies on transparent time series models. To address this gap, we propose a novel transparent neural network model for time series called Generalized Additive Time Series Model (GATSM). GATSM consists of two parts: 1) independent feature networks to learn feature representations, and 2) a transparent temporal module to learn temporal patterns across different time steps using the feature representations. This structure allows GATSM to effectively capture temporal patterns and handle varying-length time series while preserving transparency. Empirical experiments show that GATSM significantly outperforms existing generalized additive models and achieves comparable performance to black-box time series models, such as recurrent neural networks and Transformer. In addition, we demonstrate that GATSM finds interesting patterns in time series.
title Transparent Networks for Multivariate Time Series
topic Machine Learning
Computers and Society
url https://arxiv.org/abs/2410.10535