Semantic Code Repair using Neuro-Symbolic Transformation Networks

TL;DR

This paper introduces a neuro-symbolic neural network architecture for semantic code repair that predicts bug fixes without relying on test cases, achieving significantly higher accuracy than previous models.

Contribution

The work presents a novel neural network architecture with shared encoding and specialized scoring modules for semantic code repair, trained on real-world code with synthetic bugs.

Findings

Achieves 41% exact repair prediction accuracy with a single guess

Outperforms sequence-to-sequence models which have 13% accuracy

Demonstrates effectiveness on real-world GitHub bug dataset

Abstract

We study the problem of semantic code repair, which can be broadly defined as automatically fixing non-syntactic bugs in source code. The majority of past work in semantic code repair assumed access to unit tests against which candidate repairs could be validated. In contrast, the goal here is to develop a strong statistical model to accurately predict both bug locations and exact fixes without access to information about the intended correct behavior of the program. Achieving such a goal requires a robust contextual repair model, which we train on a large corpus of real-world source code that has been augmented with synthetically injected bugs. Our framework adopts a two-stage approach where first a large set of repair candidates are generated by rule-based processors, and then these candidates are scored by a statistical model using a novel neural network architecture which we refer to as Share, Specialize, and Compete. Specifically, the architecture (1) generates a shared encoding of the source code using an RNN over the abstract syntax tree, (2) scores each candidate repair using specialized network modules, and (3) then normalizes these scores together so they can compete against one another in comparable probability space. We evaluate our model on a real-world test set gathered from GitHub containing four common categories of bugs. Our model is able to predict the exact correct repair 41\% of the time with a single guess, compared to 13\% accuracy for an attentional sequence-to-sequence model.