Title :
A step towards a new test for learnability of machine learning
Author :
Inazumi, Hiroshige ; Tokiwa, Kin-ichiro ; Holte, Robert C.
Author_Institution :
Coll. of Sci. & Eng., Aoyama Gakuin Univ., Tokyo, Japan
Abstract :
Based on the PAC (probably approximately correct) learning, a new test for learnability is proposed from the viewpoint of rate distortion theory. The criterion depends on the potential property of concept classes, which shows the relationship between sample complexity and accuracy
Keywords :
learning (artificial intelligence); learning systems; rate distortion theory; testing; PAC learning; learnability test; machine learning; probably approximately correct learning; rate distortion theory; sample accuracy; sample complexity; Communication effectiveness; Entropy; Information analysis; Information theory; Machine learning; Machine learning algorithms; Rate-distortion; Sufficient conditions; Testing; Virtual colonoscopy;
Conference_Titel :
Systems, Man, and Cybernetics, 1994. Humans, Information and Technology., 1994 IEEE International Conference on
Conference_Location :
San Antonio, TX
Print_ISBN :
0-7803-2129-4
DOI :
10.1109/ICSMC.1994.399822
