Explainable artificial intelligence (XAI): Why, when, and how?

Machine learning models are rapidly conquering uncharted ground by proving themselves to be better than the existing manual or software solutions. This has also given rise to a demand for explainable artificial intelligence (XAI), which enables humans to understand the decisions made by the machine learning model. The need for EAI may stem from a number of reasons, including legal and social considerations or the desire to improve the acceptance and adoption of the machine learning model. Likewise, the extent of explainability desired may vary with the aforementioned reasons and the application domain, such as finance, defense, legal, and medical.

XAI is achieved by choosing machine learning techniques such as decision trees that lend themselves well to explainability. However, these often compromise accuracy or require additional effort to develop a secondary machine model to explain the decisions of the primary model. Essentially this leads to a choice between the desired levels of explainability, accuracy, and development cost.

Mridul Mishra discusses the need for explainability and its current state. Mridul then provides a framework for that can be used to analyze and communicate about the choices related to EAI and make the decisions that can be used to provide the best EAI solution for the problem in hand.