Program Synthesis is "Just" Another Learning Problem

Leona F. Fass

We view processes of program synthesis as behavior realization problems, within the scope of computational learning theory. If potential program behavior could be finitely, completely specified, we believe we could infer and verify programs to be correct. But realistically lacking such complete specifications, we support current research in synthesis which we believe produces finite-state correct-program approximations.


This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.