Testing Neural Program Analyzers

Md Rafiqul Islam Rabin, Ke Wang, Mohammad Amin Alipour
8/25/2019
cs.LGcs.PLcs.SE

Official code: mdrafiqulrabin/tnpa-framework (3 stars, last commit 2020-07-28).

Abstract

Deep neural networks have been increasingly used in software engineering and program analysis tasks. They usually take a program and make some predictions about it, e.g., bug prediction. We call these models neural program analyzers. The reliability of neural programs can impact the reliability of the encompassing analyses. In this paper, we describe our ongoing efforts to develop effective techniques for testing neural programs. We discuss the challenges involved in developing such tools and our future plans. In our preliminary experiment on a neural model recently proposed in the literature, we found that the model is very brittle, and simple perturbations in the input can cause the model to make mistakes in its prediction.

View on arXivView PDF

Code Implementations(1)

mdrafiqulrabin/tnpa-frameworkOfficial100%

ASE(LBR)-2019: Testing Neural Program Analyzers

30Jun 5, 20206 years ago

Similar Papers

NeuroCodeBench: a plain C neural network benchmark for software verification

83% match1 repo72023

Automated Repair of AI Code with Large Language Models and Formal Verification

82% match2024

Toward Transparent AI: A Survey on Interpreting the Inner Structures of Deep Neural Networks

81% match2022

A Survey on Uncertainty Quantification Methods for Deep Learning

81% match2023

NeuroX: A Toolkit for Analyzing Individual Neurons in Neural Networks

81% match1 repo1092018

Path Analysis for Effective Fault Localization in Deep Neural Networks

81% match1 repo2023

Cite this paper

@article{rabin2019testing,
  title  = {Testing Neural Program Analyzers},
  author = {Md Rafiqul Islam Rabin and Ke Wang and Mohammad Amin Alipour},
  year   = {2019},
  eprint = {1908.10711},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1908.10711}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1908.10711.svg)](https://codeofpaper.com/paper/1908.10711)

Discussion