Work

Explaining the Black Box: Interpretable Machine Learning Models in High-Stakes AI Systems

Year: 2024
Type: article
Abstract: In the advent of artificial intelligence (AI) permeating critical sectors such as healthcare, finance, and criminal justice, the demand for interpretability in machine learning (ML) models has grown significantly. This paper investigates state-of-the-art methods for interpretable ML and their application in high-stakes AI systems. We analyze the trade-offs between model accuracy and interpretability, review prominent interpretability techniques, and propose a framework for integrating these methods into high-stakes environments. Our findings suggest that combining human-centric design principles with interpretable ML techniques can mitigate risks and enhance trust in AI systems.
Language: English
Open Access status: diamond
OpenAlex ID: W7217099447
FWCI:
Cites:
Cited by:
Related to: 10