Explaining black boxes with a SMILE: Statistical Model-agnostic Interpretability with Local Explanations

Aslansefat, Koorosh; Hashemian, Mojgan; Walker, Martin; Akram, Mohammed Naveed; Sorokos, Ioannis; Papadopoulos, Yiannis

Abstract

Machine learning is currently undergoing an explosion in capability, popularity, and sophistication. However, one of the major barriers to widespread acceptance of machine learning (ML) is trustworthiness: most ML models operate as black boxes, their inner workings opaque and mysterious, and it can be difficult to trust their conclusions without understanding how those conclusions are reached. Explainability is therefore a key aspect of improving trustworthiness: the ability to better understand, interpret, and anticipate the behaviour of ML models. To this end, we propose SMILE, a new method that builds on previous approaches by making use of statistical distance measures to improve explainability while remaining applicable to a wide range of input data domains.

Citation

Aslansefat, K., Hashemian, M., Walker, M., Akram, M. N., Sorokos, I., & Papadopoulos, Y. (2023). Explaining black boxes with a SMILE: Statistical Model-agnostic Interpretability with Local Explanations. IEEE Software, https://doi.org/10.1109/MS.2023.3321282

• Journal Article Type: Article
• Acceptance Date: Sep 20, 2023
• Online Publication Date: Oct 4, 2023
• Publication Date: 2023
• Deposit Date: Oct 10, 2023
• Publicly Available Date: Nov 13, 2023
• Journal: IEEE Software
• Print ISSN: 0740-7459
• Electronic ISSN: 1937-4194
• Publisher: Institute of Electrical and Electronics Engineers
• Peer Reviewed: Peer Reviewed
• DOI: https://doi.org/10.1109/MS.2023.3321282
• Keywords: Closed box; Perturbation methods; Predictive models; Gaussian distribution; Data models; Machine learning; Training
• Public URL: https://hull-repository.worktribe.com/output/4415493
• Publisher URL: https://ieeexplore.ieee.org/abstract/document/10272255