Leveraging Language to Learn Program Abstractions and Search Heuristics. (arXiv:2106.11053v3 [cs.LG] UPDATED)

Web: http://arxiv.org/abs/2106.11053

Inductive program synthesis, or inferring programs from examples of desired
behavior, offers a general paradigm for building interpretable, robust, and
generalizable machine learning systems. Effective program synthesis depends on
two key ingredients: a strong library of functions from which to build
programs, and an efficient search strategy for finding programs that solve a
given task. We introduce LAPS (Language for Abstraction and Program Search), a
technique for using natural language annotations to guide joint learning of
libraries and neurally-guided search …

arxiv heuristics language search