Learning to Combine Per-Example Solutions for Neural Program Synthesis

TL;DR

This paper introduces a novel two-stage neural approach for program synthesis that combines per-example solutions using a multi-head attention mechanism, significantly improving success rates over existing methods.

Contribution

It proposes the Cross Aggregator neural network module to effectively combine per-example solutions for improved program synthesis.

Findings

Significant success rate improvements over PCCoder.

Effective combination of per-example solutions using attention.

Robust performance across different program lengths.

Abstract

The goal of program synthesis from examples is to find a computer program that is consistent with a given set of input-output examples. Most learning-based approaches try to find a program that satisfies all examples at once. Our work, by contrast, considers an approach that breaks the problem into two stages: (a) find programs that satisfy only one example, and (b) leverage these per-example solutions to yield a program that satisfies all examples. We introduce the Cross Aggregator neural network module based on a multi-head attention mechanism that learns to combine the cues present in these per-example solutions to synthesize a global solution. Evaluation across programs of different lengths and under two different experimental settings reveal that when given the same time budget, our technique significantly improves the success rate over PCCoder [Zohar et. al 2018] and other ablation baselines. The code, data and trained models for our work can be found at https://github.com/shrivastavadisha/N-PEPS.