Invited Paper: What is AI Software Testing? and Why
Jerry Zeyu Gao, Chuanqi Tao, Dou Jie, Shengqiang Lu · 2019
With the fast advance of artificial intelligence technology and data-driven machine learning techniques, building high-quality AI-based software in different application domains is becoming a very hot research topic in both academic and industry communities. Today, many machine learning models and artificial technologies have been developed to build smart application systems based on multimedia inputs to achieve intelligent functional features, such as recommendation, object detection, classification, and prediction, natural language processing and translation, and so on. This brings strong demand in quality validation and assurance for AI software systems. Current research work seldom discusses AI software testing questions, challenges, and validation approaches with clear quality requirements and criteria. This paper focuses on AI software quality validation, including validation focuses, features, and process, and potential testing approaches. Moreover, it presents a test process and a classification-based test modeling for AI classification function testing. Finally, it discusses the challenges, issues, and needs in AI software testing.