Leveraging Grammar and Reinforcement Learning for Neural Program Synthesis

TL;DR

This paper introduces a neural program synthesis method that combines grammar constraints and reinforcement learning to generate syntactically correct and semantically accurate programs, especially effective with limited training data.

Contribution

It proposes a novel training approach that explicitly maximizes the likelihood of semantically correct and syntactically valid programs using reinforcement learning and grammar-based constraints.

Findings

Improved accuracy in program synthesis with limited data

Effective enforcement of syntax correctness during training

Enhanced semantic correctness of generated programs

Abstract

Program synthesis is the task of automatically generating a program consistent with a specification. Recent years have seen proposal of a number of neural approaches for program synthesis, many of which adopt a sequence generation paradigm similar to neural machine translation, in which sequence-to-sequence models are trained to maximize the likelihood of known reference programs. While achieving impressive results, this strategy has two key limitations. First, it ignores Program Aliasing: the fact that many different programs may satisfy a given specification (especially with incomplete specifications such as a few input-output examples). By maximizing the likelihood of only a single reference program, it penalizes many semantically correct programs, which can adversely affect the synthesizer performance. Second, this strategy overlooks the fact that programs have a strict syntax that can be efficiently checked. To address the first limitation, we perform reinforcement learning on top of a supervised model with an objective that explicitly maximizes the likelihood of generating semantically correct programs. For addressing the second limitation, we introduce a training procedure that directly maximizes the probability of generating syntactically correct programs that fulfill the specification. We show that our contributions lead to improved accuracy of the models, especially in cases where the training data is limited.