English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Convolutional neural networks (CNNs) have recently attracted great attentionin geoscience due to their ability to capture non-linear system behavior andextract predictive spatiotemporal patterns. Given their black-box naturehowever, and the importance of prediction explainability, methods ofexplainable artificial intelligence (XAI) are gaining popularity as a means toexplain the CNN decision-making strategy. Here, we establish an intercomparisonof some of the most popular XAI methods and investigate their fidelity inexplaining CNN decisions for geoscientific applications. Our goal is to raiseawareness of the theoretical limitations of these methods and gain insight intothe relative strengths and weaknesses to help guide best practices. Theconsidered XAI methods are first applied to an idealized attribution benchmark,where the ground truth of explanation of the network is known a priori, to helpobjectively assess their performance. Secondly, we apply XAI to aclimate-related prediction setting, namely to explain a CNN that is trained topredict the number of atmospheric rivers in daily snapshots of climatesimulations. Our results highlight several important issues of XAI methods(e.g., gradient shattering, inability to distinguish the sign of attribution,ignorance to zero input) that have previously been overlooked in our field and,if not considered cautiously, may lead to a distorted picture of the CNNdecision-making strategy. We envision that our analysis will motivate furtherinvestigation into XAI fidelity and will help towards a cautious implementationof XAI in geoscience, which can lead to further exploitation of CNNs and deeplearning for prediction problems.

Investigating the fidelity of explainable artificial intelligence methods for applications of convolutional neural networks in geoscience