A Review of Intelligent Design for Test Based on Machine Learning

In recent years, with the development of post-Moore era, integrated circuit architecture is complex and test cost is increasing. How to apply efficient, reliable, fast and sustainable ideas to integrated circuit design has become a research hotspot in this field. Machine Learning, by virtue of its ability to quickly process abundant data information, shines in the field of Design For Test, which aims to improve the reliability of integrated circuit products and reduce design costs. In this paper, we introduce the application of Machine Learning in scan compression, Automatic Test Pattern Generation algorithm, the Test Pattern Generator in Built-In-Self- Test and test power. In addition, we prospect the foreground application of Machine Learning application in scan test, Automatic Test Pattern Generation, Built-In-Self- Test and test power.