white-box model

Terms from Artificial Intelligence: humans at the heart of algorithms

White-box models expose their internal structues, paraemters and processes, in ways that are understandable to humans. The latter is crucial as deep neural networks, for example, can in principle, interogated; however, without using techniques of explainable AI, the billions of weighst are totally in comprehensible, even for experts.

Used on pages 523, 528

Also known as white-box learning