plop: Probabilistic Learning of Programs

By Moshe Looks, Google Research

Cross-posted with the Google Research blog

Traditional machine learning systems work with relatively flat, uniform data representations, such as feature vectors, time-series, and probabilistic context-free grammars. However, reality often presents us with data which are best understood in terms of relations, types, hierarchies, and complex functional forms. The best representational scheme we computer scientists have for coping with this sort of complexity is computer programs. Yet there are comparatively few machine learning methods that operate on programmatic representations, due to the extreme combinatorial explosions involved and the semantic complexities of programs.

The plop project, launched early Monday, November 10th, is unusual in addressing these issues directly - its long-goals are quite ambitious! Plop is being implemented in Common Lisp, an equally unusual programming language that is uniquely suited to constructing and transforming complex programmatic representations.