One Explanation Does Not Fit All: A Toolkit and Taxonomy of AI Explainability Techniques

TL;DR

This paper introduces AI Explainability 360, an open-source toolkit with diverse methods and a taxonomy to help stakeholders understand AI decisions, addressing varied explanation needs across society.

Contribution

It presents a comprehensive toolkit and taxonomy for AI explainability, facilitating navigation and selection of methods for different stakeholder requirements.

Findings

Includes eight explainability methods and two evaluation metrics.

Provides an extensible architecture aligned with the AI modeling pipeline.

Offers tutorials and interactive demos for diverse audiences.

Abstract

As artificial intelligence and machine learning algorithms make further inroads into society, calls are increasing from multiple stakeholders for these algorithms to explain their outputs. At the same time, these stakeholders, whether they be affected citizens, government regulators, domain experts, or system developers, present different requirements for explanations. Toward addressing these needs, we introduce AI Explainability 360 (http://aix360.mybluemix.net/), an open-source software toolkit featuring eight diverse and state-of-the-art explainability methods and two evaluation metrics. Equally important, we provide a taxonomy to help entities requiring explanations to navigate the space of explanation methods, not only those in the toolkit but also in the broader literature on explainability. For data scientists and other users of the toolkit, we have implemented an extensible software architecture that organizes methods according to their place in the AI modeling pipeline. We also discuss enhancements to bring research innovations closer to consumers of explanations, ranging from simplified, more accessible versions of algorithms, to tutorials and an interactive web demo to introduce AI explainability to different audiences and application domains. Together, our toolkit and taxonomy can help identify gaps where more explainability methods are needed and provide a platform to incorporate them as they are developed.