TIMING: Temporality-Aware Integrated Gradients for Time Series Explanation

TL;DR

TIMING is a novel method that improves the explanation of time series predictions by incorporating temporal awareness into Integrated Gradients, leading to more accurate identification of significant points compared to existing methods.

Contribution

The paper introduces TIMING, a temporality-aware extension of Integrated Gradients, along with new evaluation metrics for better assessment of time series explanations.

Findings

TIMING outperforms existing XAI methods on synthetic and real-world benchmarks.

New metrics CPD and CPP effectively evaluate attribution methods for positive and negative impacts.

Conventional IG performs well under new metrics but needs temporal adaptation for optimal results.

Abstract

Recent explainable artificial intelligence (XAI) methods for time series primarily estimate point-wise attribution magnitudes, while overlooking the directional impact on predictions, leading to suboptimal identification of significant points. Our analysis shows that conventional Integrated Gradients (IG) effectively capture critical points with both positive and negative impacts on predictions. However, current evaluation metrics fail to assess this capability, as they inadvertently cancel out opposing feature contributions. To address this limitation, we propose novel evaluation metrics-Cumulative Prediction Difference (CPD) and Cumulative Prediction Preservation (CPP)-to systematically assess whether attribution methods accurately identify significant positive and negative points in time series XAI. Under these metrics, conventional IG outperforms recent counterparts. However, directly applying IG to time series data may lead to suboptimal outcomes, as generated paths ignore temporal relationships and introduce out-of-distribution samples. To overcome these challenges, we introduce TIMING, which enhances IG by incorporating temporal awareness while maintaining its theoretical properties. Extensive experiments on synthetic and real-world time series benchmarks demonstrate that TIMING outperforms existing time series XAI baselines. Our code is available at https://github.com/drumpt/TIMING.