“The aim of AI is to train machines to do some of the work that people were needed to do previously,” said Lei. “The training process requires a large amount of labeled data. It is time intensive and there are significant labor costs in collecting and labeling all that data. Few-shot learning can be valuable in forwarding research because it reduces the training cost by using less labeled data to get the same – and sometimes even greater – accuracy in training results.
Lei has collaborated on two published papers that incorporate her current work: Robust Regression via Heuristic Corruption Thresholding and Its Adaptive Estimation Variation,” ACM Transactions on Knowledge Discovery from Data (TKDD) 2019; and “Robust Regression via Online Feature Selection under Adversarial Data Corruption,” proceedings of the IEEE International Conference on Data Mining (ICDM), Singapore, 2018.
Lei holds bachelor’s and master’s degrees in software engineering from Beihang University in China. The opportunity to work with professors who are expert in their fields, other talented students, and the northern Virginia location attracted her to Virginia Tech and the Discovery Analytics Center.
“I think the best thing about DAC is its abundance of academic resources,” Lei said. “I really enjoy working with everyone there. Dr. Lu provides a lot of support for my research and I have also learned from my lab members. They are very nice and helpful, always willing to offer suggestions whenever I have encountered a problem.”
Lei will spend the summer months engaged in her research at DAC.
Her previous interest in spatial temporal data mining, which included resident travel pattern analysis, is reflected in a collaborative paper, “Forecasting car rental demand based temporal and spatial travel patterns,” in 2017 IEEE Ubiquitous Intelligence, Cloud and Big Data Computing, Internet of People and Smart City Innovation.
When she has free time, Lei enjoys cooking, baking, traveling, and photography.
She is projected to graduate in May 2022.