A Machine Learning Framework for Programming by Example

Aditya Menon, Omer Tamuz, Sumit Gulwani, Butler Lampson, Adam Kalai
; Proceedings of the 30th International Conference on Machine Learning, PMLR 28(1):187-195, 2013.

Abstract

Learning programs is a timely and interesting challenge. In Programming by Example (PBE), a system attempts to infer a program from input and output examples alone, by searching for a composition of some set of base functions. We show how machine learning can be used to speed up this seemingly hopeless search problem, by learning weights that relate textual features describing the provided input-output examples to plausible sub-components of a program. This generic learning framework lets us address problems beyond the scope of earlier PBE systems. Experiments on a prototype implementation show that learning improves search and ranking on a variety of text processing tasks found on help forums.

Cite this Paper


BibTeX
@InProceedings{pmlr-v28-menon13, title = {A Machine Learning Framework for Programming by Example}, author = {Aditya Menon and Omer Tamuz and Sumit Gulwani and Butler Lampson and Adam Kalai}, booktitle = {Proceedings of the 30th International Conference on Machine Learning}, pages = {187--195}, year = {2013}, editor = {Sanjoy Dasgupta and David McAllester}, volume = {28}, number = {1}, series = {Proceedings of Machine Learning Research}, address = {Atlanta, Georgia, USA}, month = {17--19 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v28/menon13.pdf}, url = {http://proceedings.mlr.press/v28/menon13.html}, abstract = {Learning programs is a timely and interesting challenge. In Programming by Example (PBE), a system attempts to infer a program from input and output examples alone, by searching for a composition of some set of base functions. We show how machine learning can be used to speed up this seemingly hopeless search problem, by learning weights that relate textual features describing the provided input-output examples to plausible sub-components of a program. This generic learning framework lets us address problems beyond the scope of earlier PBE systems. Experiments on a prototype implementation show that learning improves search and ranking on a variety of text processing tasks found on help forums.} }
Endnote
%0 Conference Paper %T A Machine Learning Framework for Programming by Example %A Aditya Menon %A Omer Tamuz %A Sumit Gulwani %A Butler Lampson %A Adam Kalai %B Proceedings of the 30th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2013 %E Sanjoy Dasgupta %E David McAllester %F pmlr-v28-menon13 %I PMLR %J Proceedings of Machine Learning Research %P 187--195 %U http://proceedings.mlr.press %V 28 %N 1 %W PMLR %X Learning programs is a timely and interesting challenge. In Programming by Example (PBE), a system attempts to infer a program from input and output examples alone, by searching for a composition of some set of base functions. We show how machine learning can be used to speed up this seemingly hopeless search problem, by learning weights that relate textual features describing the provided input-output examples to plausible sub-components of a program. This generic learning framework lets us address problems beyond the scope of earlier PBE systems. Experiments on a prototype implementation show that learning improves search and ranking on a variety of text processing tasks found on help forums.
RIS
TY - CPAPER TI - A Machine Learning Framework for Programming by Example AU - Aditya Menon AU - Omer Tamuz AU - Sumit Gulwani AU - Butler Lampson AU - Adam Kalai BT - Proceedings of the 30th International Conference on Machine Learning PY - 2013/02/13 DA - 2013/02/13 ED - Sanjoy Dasgupta ED - David McAllester ID - pmlr-v28-menon13 PB - PMLR SP - 187 DP - PMLR EP - 195 L1 - http://proceedings.mlr.press/v28/menon13.pdf UR - http://proceedings.mlr.press/v28/menon13.html AB - Learning programs is a timely and interesting challenge. In Programming by Example (PBE), a system attempts to infer a program from input and output examples alone, by searching for a composition of some set of base functions. We show how machine learning can be used to speed up this seemingly hopeless search problem, by learning weights that relate textual features describing the provided input-output examples to plausible sub-components of a program. This generic learning framework lets us address problems beyond the scope of earlier PBE systems. Experiments on a prototype implementation show that learning improves search and ranking on a variety of text processing tasks found on help forums. ER -
APA
Menon, A., Tamuz, O., Gulwani, S., Lampson, B. & Kalai, A.. (2013). A Machine Learning Framework for Programming by Example. Proceedings of the 30th International Conference on Machine Learning, in PMLR 28(1):187-195

Related Material