Reassessing Machine Learning Techniques for Electrocatalyst Design: A Call for Robust Methodologies

Keywords

electrocatalysts / feature selection / machine learning / SHAP / Spearman's correlation

References

[1] Y. Gu, H. Zhang, Z. Xu, et al., “Design and Application of Electrocatalyst Based on Machine Learning,” Interdisciplinary Materials 4 (2025): 456-479, https://doi.org/10.1002/idm2.12249.

[2] L. Wu, “A Review of the Transition From Shapley Values and SHAP Values to RGE,” Statistics 1 (2025): 1-23, https://doi.org/10.1080/02331888.2025.2487853.

[3] B. Bilodeau, N. Jaques, P. W. Koh, and B. Kim, “Impossibility Theorems for Feature Attribution,” Proceedings of the National Academy of Sciences of the United States of America 121, no. 2 (2024): e2304406120, https://doi.org/10.1073/pnas.2304406120.

[4] X. Huang and J. Marques-Silva, “On the Failings of Shapley Values for Explainability,” International Journal of Approximate Reasoning 171 (2024): 109112, https://doi.org/10.1016/j.ijar.2023.109112.

[5] D. Hooshyar and Y. Yang, “Problems With SHAP and LIME in Interpretable AI for Education: A Comparative Study of Post-Hoc Explanations and Neural-Symbolic Rule Extraction,” IEEE Access 12 (2024): 137472-137490, https://doi.org/10.1109/ACCESS.2024.3463948.

RIGHTS & PERMISSIONS

2025 The Author(s). Interdisciplinary Materials published by Wuhan University of Technology and John Wiley & Sons Australia, Ltd.