ControlFlag: a self-supervised idiosyncratic pattern detection system for software control structures (2020-11-06T00:00:00.000000Z)

TL;DR

ControlFlag is presented, a self-supervised MP system that aims to improve debugging by attempting to detect idiosyncratic pattern violations in software control structures and suggests possible corrections in the event an anomalous pattern is detected.

Abstract

Software debugging has been shown to utilize upwards of half of developers’ time. Yet, machine programming (MP), the field concerned with the automation of software (and hardware) development, has recently made strides in both research and production-quality automated debugging systems. In this paper we present ControlFlag, a self-supervised MP system that aims to improve debugging by attempting to detect idiosyncratic pattern violations in software control structures. ControlFlag also suggests possible corrections in the event an anomalous pattern is detected. We present ControlFlag’s design and provide an experimental evaluation and analysis of its efficacy in identifying potential programming errors in production-quality software. As a first concrete evidence towards improving software quality, ControlFlag has already found an anomaly in CURL that has been acknowledged and fixed by its developers. We also discuss future extensions of ControlFlag.

Authors

Niranjan Hasabnis

1 papers

Justin Emile Gottschlich

1 papers

TL;DR

Abstract

Authors

References57 items

Predictive data locality optimization for higher-order tensor computations

Feedback-driven semi-supervised synthesis of program transformations

Guided pattern mining for API misuse detection by change-based code analysis

Synthesize, Execute and Debug: Learning to Repair for Neural Program Synthesis

MISIM: An End-to-End Neural Code Similarity System

MISIM: A Novel Code Similarity System.

Graph-based, Self-Supervised Program Repair from Diagnostic Feedback

Hoppity: Learning Graph Transformations to Detect and Fix Bugs in Programs

Learned garbage collection

Learning to Represent Programs with Property Signatures

Automated program repair

An Abstraction-Based Framework for Neural Network Verification

Mining Patterns in Source Code Using Tree Mining Algorithms

Learning Fitness Functions for Machine Programming

Pythia: AI-assisted Code Completion System

Program Synthesis and Semantic Parsing with Learned Code Idioms

Learning Programmatic Idioms for Scalable Semantic Parsing

Neural Program Repair by Jointly Learning to Localize and Repair

MLSys: The New Frontier of Machine Learning Systems

Automatically identifying code features for software defect prediction: Using AST N-grams

Aroma: code recommendation via structural code search

Mining Semantic Loop Idioms

DeepBugs: a learning approach to name-based bug detection

The three pillars of machine programming

Learning to Represent Programs with Graphs

A Zero-Positive Learning Approach for Diagnosing Software Performance Regressions

AI programmer: autonomously creating software programs using genetic algorithms

Extracting instruction semantics via symbolic execution of code generators

Verified lifting of stencil computations

Lifting Assembly to Intermediate Representation: A Novel Approach Leveraging Compilers

Automated Correction for Syntax Errors in Programming Assignments using Recurrent Neural Networks

Parameter-free probabilistic API mining across GitHub

Mining idioms from source code

Concurrent predicates: A debugging technique for every parallel programmer

Visualizing transactional memory

Mining Source Code for Structural Regularities

Code template inference using language models

Graph-based mining of multiple object usage patterns

Introduction to Algorithms, third edition

Mining API Error-Handling Specifications from Source Code

Parseweb: a programmer assistant for reusing open source code on the web

MAPO: mining API usages from open source repositories

DynaMine: finding common error patterns by mining software revision histories

Efficiently mining frequent trees in a forest: algorithms and applications

Mining Temporal Specifications for Error Detection

A Critique of Software Defect Prediction Models

Finding Idioms in Source Code Using Subtree Counting Techniques

Wolf Garbe

Re: Potential Confusion in http_proxy.c and a Recommendation

http_proxy: Use Enum with State Names for ‘keepon’

A Language-Parametric Modular Framework for Mining Idiomatic Code Patterns

Ioannis Mitliagkas, Rajat Monga, Derek Murray, Kunle Olukotun, Dimitris Papailiopoulos, Gennady Pekhimenko, Theodoros Rekatsinas

An Incremental Parsing System for Programming Tools

ControlFlag: A Self-Supervised Idiosyncratic Pattern Detection System for Software

How to Write a Spelling Corrector, 2016

Reversible Debugging Software — Quantify the Time and Cost Saved Using Reversible Debuggers

Standard Gotchas Subtleties in the Verilog and System Verilog Standards that Every Engineer Should Know

Field of Study

Journal Information

Name