Research on Analog Integrated Circuit Test Parameter Set Reduction Based on XGBoost

As the scale of integrated circuits continues to increase and their test cost increases with test time, how to optimize the test parameters is an important topic. In analog integrated circuits, the implicit dependency among test parameters makes it possible to apply the XGBoost technique based on decision trees in machine learning to optimize the test parameters. In this paper, an optimization algorithm is proposed based on the XGBoost decision tree model. By modeling the representational relationships of each test parameter in the historical test data set, the list of those to be optimized is obtained according to the descending order of the escape rate in the prediction results. According to this list, the test parameters to be deleted are selected in turn, the prediction results of the remaining test parameters on those test parameters are obtained, and the escape rate after screening out the target parameters is evaluated, and the test parameters are optimized based on this list to reduce the test time and test cost.