A Two-Stage Neural Network Classifier for Data Imbalanced Problem with Application to Hotspot Detection
TimeTuesday, December 7th11:30am - 11:52am PST
Event Type
Research Manuscript
Virtual Programs
Presented In-Person
Physical Design and Verification, Lithography and DFM
DescriptionThe data imbalance problem often occurs in nanometer VLSI applications, where normal cases far outnumber error ones. Existing methods focus on improving the quality of minority classes while causing quality deterioration on majority ones. This paper proposes a two-stage classifier to handle the data imbalance problem. We first develop an iterative neural network framework to reduce false alarms. Then the oversampling method on a final classification network is applied to predict the two classes better. Compared with state-of-the-art imbalanced data handling methods, experimental results show that our two-stage classification method achieves the best prediction accuracy and significantly reduces false alarms.