Abstract:Learning distributed representations of source code has been a challenging task for machine learning models. Earlier works treated programs as text so that natural language methods can be readily applied. Unfortunately, such approaches do not capitalize on the rich structural information possessed by source code. Of late, Graph Neural Network (GNN) was proposed to learn embeddings of programs from their graph representations. Due to the homogeneous (i.e. do not take advantage of the program-specific graph characteristics) and expensive (i.e. require heavy information exchange among nodes in the graph) message-passing procedure, GNN can suffer from precision issues, especially when dealing with programs rendered into large graphs. In this paper, we present a new graph neural architecture, called Graph Interval Neural Network (GINN), to tackle the weaknesses of the existing GNN. Unlike the standard GNN, GINN generalizes from a curated graph representation obtained through an abstraction method designed to aid models to learn. In particular, GINN focuses exclusively on intervals (generally manifested in looping construct) for mining the feature representation of a program, furthermore, GINN operates on a hierarchy of intervals for scaling the learning to large graphs. We evaluate GINN for two popular downstream applications: variable misuse prediction and method name prediction. Results show in both cases GINN outperforms the state-of-the-art models by a comfortable margin. We have also created a neural bug detector based on GINN to catch null pointer deference bugs in Java code. While learning from the same 9,000 methods extracted from 64 projects, GINN-based bug detector significantly outperforms GNN-based bug detector on 13 unseen test projects. Next, we deploy our trained GINN-based bug detector and Facebook Infer, arguably the state-of-the-art static analysis tool, to scan the codebase of 20 highly starred projects on GitHub. Through our manual inspection, we confirm 38 bugs out of 102 warnings raised by GINN-based bug detector compared to 34 bugs out of 129 warnings for Facebook Infer. We have reported 38 bugs GINN caught to developers, among which 11 have been fixed and 12 have been confirmed (fix pending). GINN has shown to be a general, powerful deep neural network for learning precise, semantic program embeddings.

Graphsearchnet: Enhancing gnns via capturing global dependencies for semantic code search

GraphSearchNet: Enhancing GNNs via capturing global dependencies for semantic code search.(2023)

Utilizing Edge Attention in Graph-Based Code Search

deGraphCS : Embedding Variable-based Flow Graph for Neural Code Search

<scp>de</scp> G <scp>raph</scp> CS: Embedding Variable-based Flow Graph for Neural Code Search

Survey of Code Search Based on Deep Learning

Deep Semantics-Enhanced Neural Code Search

Deep Graph Matching and Searching for Semantic Code Retrieval

Rethinking Graph Neural Network Search from Message-passing

How to Better Utilize Code Graphs in Semantic Code Search?

Rethinking Graph Neural Architecture Search From Message-Passing.

Toward Interpretable Graph Tensor Convolution Neural Network for Code Semantics Embedding.

Learning semantic program embeddings with graph interval neural network

Graph Embedding Based Code Search in Software Project

Search for Deep Graph Neural Networks

Node-dependent Semantic Search over Heterogeneous Graph Neural Networks

Code classification with graph neural networks: Have you ever struggled to make it work?

GLSearch: Maximum Common Subgraph Detection Via Learning to Search

Software Vulnerability Detection via Deep Learning over Disaggregated Code Graph Representation

Search to Capture Long-range Dependency with Stacking GNNs for Graph Classification.

Cosea: Convolutional code search with layer-wise attention