explanation

Terms from Artificial Intelligence: humans at the heart of algorithms

Page numbers are for draft copy at present; they will be replaced with correct numbers when final book is formatted. Chapter numbers are correct and will not change now.

Sometimes we only care about accuracy of an AI system, does it give the right answer. However, often we also care about why it gave the answer, that is we need an explanation. This is particularly important in complex black-box machine learning and this has give rise to the study of explainable AI. Alternatiebly, explanations may be used by the AI system in explanation-based learning. Human explanations may also be used as part of the creation of AI systems, for example in knowledge elicitation, or during ongoing interactions with systems that continue to learn in use.

Used in Chap. 20: page 335