explanation-based learning

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.

Explanation-based learning tries to buiild an explanation of a situation in order to generalise its learning. For example, if a cup has just fallen to the floor, a robot may work out that this happened because it bumped the table as it moved, and therefore create new rules to slow down when close to tables with objects near the edge. Without unpacking the explanation, the robot might simply learn to avoid going clsoe to the particular table.

Defined on page 93

Used on Chap. 5: pages 90, 93, 103, 104, 107

Also known as EBL