Neural Guided Constraint Logic Programming For Program Synthesis

Authors:
Lisa Zhang University of Toronto
Gregory Rosenblatt University of Alabama at Birmingham
Ethan Fetaya University of Toronto
Renjie Liao University of Toronto
William Byrd University of Alabama at Birmingham
Matthew Might University of Alabama at Birmingham
Raquel Urtasun University of Toronto
Richard Zemel Vector Institute/University of Toronto

Introduction:

Synthesizing programs using example input/outputs is a classic problem in artificial intelligence.The authors present a method for solving Programming By Example (PBE) problems by using a neural model to guide the search of a constraint logic programming system called miniKanren.

Abstract:

Synthesizing programs using example input/outputs is a classic problem in artificial intelligence. We present a method for solving Programming By Example (PBE) problems by using a neural model to guide the search of a constraint logic programming system called miniKanren. Crucially, the neural model uses miniKanren's internal representation as input; miniKanren represents a PBE problem as recursive constraints imposed by the provided examples. We explore Recurrent Neural Network and Graph Neural Network models. We contribute a modified miniKanren, drivable by an external agent, available at https://github.com/xuexue/neuralkanren. We show that our neural-guided approach using constraints can synthesize programs faster in many cases, and importantly, can generalize to larger problems.

You may want to know: