SparseST: Exploiting Data Sparsity in Spatiotemporal Modeling and Prediction

TL;DR

SparseST introduces a novel approach that leverages data sparsity to create more efficient spatiotemporal models, significantly reducing computational costs while maintaining high performance, especially suitable for edge devices in complex physical systems.

Contribution

The paper proposes SparseST, a pioneering framework that exploits data sparsity in spatiotemporal modeling, and introduces a multi-objective loss to balance performance and efficiency.

Findings

Achieves reduced computational cost without sacrificing accuracy.

Provides a practical Pareto front approximation for model optimization.

Demonstrates effectiveness in real-world CPS applications.

Abstract

Spatiotemporal data mining (STDM) has a wide range of applications in various complex physical systems (CPS), i.e., transportation, manufacturing, healthcare, etc. Among all the proposed methods, the Convolutional Long Short-Term Memory (ConvLSTM) has proved to be generalizable and extendable in different applications and has multiple variants achieving state-of-the-art performance in various STDM applications. However, ConvLSTM and its variants are computationally expensive, which makes them inapplicable in edge devices with limited computational resources. With the emerging need for edge computing in CPS, efficient AI is essential to reduce the computational cost while preserving the model performance. Common methods of efficient AI are developed to reduce redundancy in model capacity (i.e., model pruning, compression, etc.). However, spatiotemporal data mining naturally requires extensive model capacity, as the embedded dependencies in spatiotemporal data are complex and hard to capture, which limits the model redundancy. Instead, there is a fairly high level of data and feature redundancy that introduces an unnecessary computational burden, which has been largely overlooked in existing research. Therefore, we developed a novel framework SparseST, that pioneered in exploiting data sparsity to develop an efficient spatiotemporal model. In addition, we explore and approximate the Pareto front between model performance and computational efficiency by designing a multi-objective composite loss function, which provides a practical guide for practitioners to adjust the model according to computational resource constraints and the performance requirements of downstream tasks.