TsSHAP: Robust model agnostic feature-based explainability for time series forecasting

TL;DR

TsSHAP is a model-agnostic, feature-based explainability method for time series forecasting that provides interpretable insights into black-box models using SHAP values, applicable at local, semi-local, and global levels.

Contribution

The paper introduces TsSHAP, a novel explainability algorithm for time series forecasting that is model-agnostic and leverages SHAP values for interpretability.

Findings

Effective explanations generated for various datasets.

Robustness of TsSHAP demonstrated through extensive experiments.

Applicable to any black-box forecasting model.

Abstract

A trustworthy machine learning model should be accurate as well as explainable. Understanding why a model makes a certain decision defines the notion of explainability. While various flavors of explainability have been well-studied in supervised learning paradigms like classification and regression, literature on explainability for time series forecasting is relatively scarce. In this paper, we propose a feature-based explainability algorithm, TsSHAP, that can explain the forecast of any black-box forecasting model. The method is agnostic of the forecasting model and can provide explanations for a forecast in terms of interpretable features defined by the user a prior. The explanations are in terms of the SHAP values obtained by applying the TreeSHAP algorithm on a surrogate model that learns a mapping between the interpretable feature space and the forecast of the black-box model. Moreover, we formalize the notion of local, semi-local, and global explanations in the context of time series forecasting, which can be useful in several scenarios. We validate the efficacy and robustness of TsSHAP through extensive experiments on multiple datasets.