Score-matching-based Structure Learning for Temporal Data on Networks

TL;DR

This paper introduces PICK, an efficient score-matching algorithm for causal structure learning in static and temporal network data, improving scalability and accuracy in complex, dependent datasets.

Contribution

The paper develops a new parent-finding subroutine that significantly speeds up score-matching-based causal discovery for both i.i.d. and temporal network data.

Findings

PICK accelerates score matching by optimizing the pruning step.

The algorithm maintains high accuracy on complex datasets.

It effectively handles spatial and temporal dependencies.

Abstract

Causal discovery is a crucial initial step in establishing causality from empirical data and background knowledge. Numerous algorithms have been developed for this purpose. Among them, the score-matching method has demonstrated superior performance across various evaluation metrics, particularly for the commonly encountered Additive Nonlinear Causal Models. However, current score-matching-based algorithms are primarily designed to analyze independent and identically distributed (i.i.d.) data. More importantly, they suffer from high computational complexity due to the pruning step required for handling dense Directed Acyclic Graphs (DAGs). To enhance the scalability of score matching, we have developed a new parent-finding subroutine for leaf nodes in DAGs, significantly accelerating the most time-consuming part of the process: the pruning step. This improvement results in an efficiency-lifted score matching algorithm, termed Parent Identification-based Causal structure learning for both i.i.d. and temporal data on networKs, or PICK. The new score-matching algorithm extends the scope of existing algorithms and can handle static and temporal data on networks with weak network interference. Our proposed algorithm can efficiently cope with increasingly complex datasets that exhibit spatial and temporal dependencies, commonly encountered in academia and industry. The proposed algorithm can accelerate score-matching-based methods while maintaining high accuracy in real-world applications.