神经网络国际验证竞赛(VNN-COMP)前三年

IF 1.1 3区计算机科学 Q4 COMPUTER SCIENCE, SOFTWARE ENGINEERING

International Journal on Software Tools for Technology Transfer Pub Date : 2023-05-30 DOI:10.1007/s10009-023-00703-4

Christopher Brix, Mark Niklas Müller, Stanley Bak, Taylor T. Johnson, Changliu Liu

{"title":"神经网络国际验证竞赛(VNN-COMP)前三年","authors":"Christopher Brix, Mark Niklas Müller, Stanley Bak, Taylor T. Johnson, Changliu Liu","doi":"10.1007/s10009-023-00703-4","DOIUrl":null,"url":null,"abstract":"Abstract This paper presents a summary and meta-analysis of the first three iterations of the annual International Verification of Neural Networks Competition (VNN-COMP), held in 2020, 2021, and 2022. In the VNN-COMP, participants submit software tools that analyze whether given neural networks satisfy specifications describing their input-output behavior. These neural networks and specifications cover a variety of problem classes and tasks, corresponding to safety and robustness properties in image classification, neural control, reinforcement learning, and autonomous systems. We summarize the key processes, rules, and results, present trends observed over the last three years, and provide an outlook into possible future developments.","PeriodicalId":14395,"journal":{"name":"International Journal on Software Tools for Technology Transfer","volume":"168 1","pages":"0"},"PeriodicalIF":1.1000,"publicationDate":"2023-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"16","resultStr":"{\"title\":\"First three years of the international verification of neural networks competition (VNN-COMP)\",\"authors\":\"Christopher Brix, Mark Niklas Müller, Stanley Bak, Taylor T. Johnson, Changliu Liu\",\"doi\":\"10.1007/s10009-023-00703-4\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Abstract This paper presents a summary and meta-analysis of the first three iterations of the annual International Verification of Neural Networks Competition (VNN-COMP), held in 2020, 2021, and 2022. In the VNN-COMP, participants submit software tools that analyze whether given neural networks satisfy specifications describing their input-output behavior. These neural networks and specifications cover a variety of problem classes and tasks, corresponding to safety and robustness properties in image classification, neural control, reinforcement learning, and autonomous systems. We summarize the key processes, rules, and results, present trends observed over the last three years, and provide an outlook into possible future developments.\",\"PeriodicalId\":14395,\"journal\":{\"name\":\"International Journal on Software Tools for Technology Transfer\",\"volume\":\"168 1\",\"pages\":\"0\"},\"PeriodicalIF\":1.1000,\"publicationDate\":\"2023-05-30\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"16\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"International Journal on Software Tools for Technology Transfer\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1007/s10009-023-00703-4\",\"RegionNum\":3,\"RegionCategory\":\"计算机科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"COMPUTER SCIENCE, SOFTWARE ENGINEERING\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal on Software Tools for Technology Transfer","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1007/s10009-023-00703-4","RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"COMPUTER SCIENCE, SOFTWARE ENGINEERING","Score":null,"Total":0}

引用次数: 16

摘要

摘要本文对2020年、2021年和2022年举行的年度国际神经网络验证竞赛(VNN-COMP)的前三次迭代进行了总结和荟萃分析。在VNN-COMP中，参与者提交软件工具来分析给定的神经网络是否满足描述其输入-输出行为的规范。这些神经网络和规范涵盖了各种各样的问题类别和任务，对应于图像分类、神经控制、强化学习和自主系统中的安全性和鲁棒性。我们总结了过去三年中观察到的关键过程、规则和结果、目前的趋势，并对未来可能的发展进行了展望。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

First three years of the international verification of neural networks competition (VNN-COMP)

Abstract This paper presents a summary and meta-analysis of the first three iterations of the annual International Verification of Neural Networks Competition (VNN-COMP), held in 2020, 2021, and 2022. In the VNN-COMP, participants submit software tools that analyze whether given neural networks satisfy specifications describing their input-output behavior. These neural networks and specifications cover a variety of problem classes and tasks, corresponding to safety and robustness properties in image classification, neural control, reinforcement learning, and autonomous systems. We summarize the key processes, rules, and results, present trends observed over the last three years, and provide an outlook into possible future developments.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

International Journal on Software Tools for Technology Transfer Computer Science-Information Systems

CiteScore

4.50

自引率

6.70%

发文量

期刊介绍： The International Journal on Software Tools for Technology Transfer (STTT) provides a forum for the discussion of all aspects of tools supporting the development of computer systems. It offers, above all, a tool-oriented link between academic research and industrial practice. Tool support for the development of reliable and correct computer-based systems is of growing importance, and a wealth of design methodologies, algorithms, and associated tools have been developed in different areas of computer science. However, each area has its own culture and terminology, preventing researchers from taking advantage of the results obtained by colleagues in other fields. Tool builders are often unaware of the work done by others, and thus unable to apply it. The situation is even more critical when considering the transfer of new technology into industrial practice.