L2GTX: From Local to Global Time Series Explanations

2026-03-13

AI Summary

Existing explainable AI methods struggle to generate global explanations for time series classification that simultaneously capture temporal dependencies and recurring cross-instance patterns, often being limited to specific models or unsuitable for temporal data. This work proposes L2GTX, the first model-agnostic framework for global time series explanation. L2GTX first produces local explanations via LOMATCE, extracting and parameterizing temporal event primitives—such as trends and extrema—and then constructs compact, human-readable, high-fidelity class-level global explanations through cross-instance clustering, redundancy reduction, and selection of representative instances. Experiments on six benchmark datasets demonstrate that L2GTX significantly enhances explanation conciseness and interpretability while maintaining stable global fidelity.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Humans and AI: Explainable AI (XAI) for Human Understanding
• Computer Vision: Interpretability, Explainability, and Transparency

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Deep learning models achieve high accuracy in time series classification, yet understanding their class-level decision behaviour remains challenging. Explanations for time series must respect temporal dependencies and identify patterns that recur across instances. Existing approaches face three limitations: model-agnostic XAI methods developed for images and tabular data do not readily extend to time series, global explanation synthesis for time series remains underexplored, and most existing global approaches are model-specific. We propose L2GTX, a model-agnostic framework that generates class-wise global explanations by aggregating local explanations from a representative set of instances. L2GTX extracts clusters of parameterised temporal event primitives, such as increasing or decreasing trends and local extrema, together with their importance scores from instance-level explanations produced by LOMATCE. These clusters are merged across instances to reduce redundancy, and an instance-cluster importance matrix is used to estimate global relevance. Under a user-defined instance selection budget, L2GTX selects representative instances that maximise coverage of influential clusters. Events from the selected instances are then aggregated into concise class-wise global explanations. Experiments on six benchmark time series datasets show that L2GTX produces compact and interpretable global explanations while maintaining stable global faithfulness measured as mean local surrogate fidelity.

Problem

• time series explanation
• global explanation
• model-agnostic XAI
• temporal dependencies
• class-level interpretation

Innovation

• time series explanation
• global explanation
• model-agnostic XAI
• temporal event primitives
• instance selection