- author = "Pavol Bielik and Veselin Raychev and Martin T. Vechev",
- title = "{PHOG:} Probabilistic Model for Code",
- booktitle = "Proceedings of the 33nd International Conference on Machine Learning, ICML 2016",
- year = "2016",
- editor = "Maria-Florina Balcan and Kilian Q. Weinberger",
- series = "JMLR Workshop and Conference Proceedings",
- volume = "48",
- pages = "2933--2942",
- address = "New York City, NY, USA",
- month = jun # " 19-24",
- publisher = "PMLR",
- keywords = "genetic algorithms, genetic programming, seeding",
- timestamp = "Wed, 29 May 2019 08:41:46 +0200",
-
biburl = "
https://dblp.org/rec/conf/icml/BielikRV16.bib",
- bibsource = "dblp computer science bibliography, https://dblp.org",
-
URL = "http://proceedings.mlr.press/v48/bielik16.html",
-
URL = "
http://proceedings.mlr.press/v48/bielik16.pdf",
- size = "10 pages",
- abstract = "We introduce a new generative model for code called probabilistic higher order grammar (PHOG). PHOG generalises probabilistic context free grammars (PCFGs) by allowing conditioning of a production rule beyond the parent non-terminal, thus capturing rich contexts relevant to programs. Even though PHOG is more powerful than a PCFG, it can be learned from data just as efficiently. We trained a PHOG model on a large JavaScript code corpus and show that it is more precise than existing models, while similarly fast. As a result, PHOG can immediately benefit existing programming tools based on probabilistic models of code.",
-
notes = "'Learning of TCOND Functions... Enumerative search is
exponential we use it only on short functions with up
to 5 instructions... The resulting functions serve as a
starting population for a follow-up genetic programming
search... We do not apply a cross-over operation in the
genetic search procedure.
Also known as \cite{pmlr-v48-bielik16}
Cited by \cite{DBLP:conf/pldi/LeeHAN18}",
