Inference Over Programs That Make Predictions

TL;DR

This paper discusses extending probabilistic programming for program induction to enable automatic synthesis that generalizes across diverse data types like text, images, and videos.

Contribution

It proposes future directions to expand probabilistic program induction for broader, more generalizable applications across various data modalities.

Findings

Builds on previous probabilistic programming work

Outlines steps for generalizing program synthesis

Aims to handle complex data types like text, images, videos

Abstract

This abstract extends on the previous work (arXiv:1407.2646, arXiv:1606.00075) on program induction using probabilistic programming. It describes possible further steps to extend that work, such that, ultimately, automatic probabilistic program synthesis can generalise over any reasonable set of inputs and outputs, in particular in regard to text, image and video data.