Heterogeneous Relational Kernel Learning

TL;DR

This paper introduces a novel Bayesian kernel embedding method for heterogeneous time series, enabling interpretable analysis and applications like clustering, pattern discovery, and anomaly detection with minimal additional computational cost.

Contribution

It extends relational kernel learning to heterogeneous data, providing interpretable embeddings and practical utility in clustering and anomaly detection.

Findings

Effective clustering of heterogeneous time series

Enhanced pattern discovery capabilities

Successful anomaly detection in diverse datasets

Abstract

Recent work has developed Bayesian methods for the automatic statistical analysis and description of single time series as well as of homogeneous sets of time series data. We extend prior work to create an interpretable kernel embedding for heterogeneous time series. Our method adds practically no computational cost compared to prior results by leveraging previously discarded intermediate results. We show the practical utility of our method by leveraging the learned embeddings for clustering, pattern discovery, and anomaly detection. These applications are beyond the ability of prior relational kernel learning approaches.