Finding the right XAI method -- A Guide for the Evaluation and Ranking of Explainable AI Methods in Climate Science

TL;DR

This paper evaluates and compares various XAI methods in climate science, focusing on properties like robustness and faithfulness, to guide researchers in selecting suitable explainability techniques for climate-related machine learning models.

Contribution

It introduces a comprehensive evaluation framework for XAI methods in climate science and benchmarks their performance across multiple explanation properties.

Findings

Integrated Gradients, LRP, and input times gradients show high robustness and faithfulness.

Sensitivity methods perform well in randomization but less in faithfulness.

Performance varies with network architecture, emphasizing task-specific evaluation.

Abstract

Explainable artificial intelligence (XAI) methods shed light on the predictions of machine learning algorithms. Several different approaches exist and have already been applied in climate science. However, usually missing ground truth explanations complicate their evaluation and comparison, subsequently impeding the choice of the XAI method. Therefore, in this work, we introduce XAI evaluation in the climate context and discuss different desired explanation properties, namely robustness, faithfulness, randomization, complexity, and localization. To this end, we chose previous work as a case study where the decade of annual-mean temperature maps is predicted. After training both a multi-layer perceptron (MLP) and a convolutional neural network (CNN), multiple XAI methods are applied and their skill scores in reference to a random uniform explanation are calculated for each property. Independent of the network, we find that XAI methods Integrated Gradients, layer-wise relevance propagation, and input times gradients exhibit considerable robustness, faithfulness, and complexity while sacrificing randomization performance. Sensitivity methods -- gradient, SmoothGrad, NoiseGrad, and FusionGrad, match the robustness skill but sacrifice faithfulness and complexity for randomization skill. We find architecture-dependent performance differences regarding robustness, complexity and localization skills of different XAI methods, highlighting the necessity for research task-specific evaluation. Overall, our work offers an overview of different evaluation properties in the climate science context and shows how to compare and benchmark different explanation methods, assessing their suitability based on strengths and weaknesses, for the specific research problem at hand. By that, we aim to support climate researchers in the selection of a suitable XAI method.