Lin, Tong received the PhD degree in Applied Mathematics from Peking University in 2001. In 2002, he joined the Key Laboratory of Machine Perception at Peking University, China, where he is currently an associate professor. From 2004 to 2005, he was an exchange scholar at Seoul National University, Korea. From 2007 to 2008, he was an exchange scholar at UCSD Moores Cancer Center, CA, USA. His research interests are machine learning theory and algorithms, with applications in medical data analysis and computational finance (quantitative trading).

He has published more than 30 peer-reviewed academic papers on international journals and conferences, including Journal of Machine Learning Research (JMLR), IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), IEEE Transactions on Image Processing, Signal Processing, Physics in Medicine and Biology (PMB), Journal of Trace Elements in Medicine and Biology (JTEMB), International Conference on Machine Learning (ICML), European Conference on Computer Vision (ECCV). In most papers he was the first author. He has four invention patents licensed.

He has several research projects including NSFC and 973 programs. His research works are summarized as follows:

1) Euler’s elastica model for supervised learning. Overfitting is a central problem in training a supervised classifier, and typical regularizers like L2 norms or kernel norms cannot guarantee the smoothness of the decision boundaries and the discriminant functions. We proposed Euler’s elastic model to regularize kernel-based classifiers, and experiments on classification and regression problems showed the improvements of this new model over traditional methods like SVM and neural networks with single hidden layer. We also prove that the proposed method has Bayes consistency. This work was published on ICML’2012 and JMLR’2015 (with 50 pages).

2) Manifold Learning. We proposed a nonlinear dimensionality reduction method based on Riemannian normal coordinates, and showed several applications to embed high-dimensional data into an intrinsic low-dimensional space. This work was published on PAMI’2008.

3) Machine learning methods for medical data analysis. We developed tumor gating and tracking methods for Image-Guided Lung Cancer Radiotherapy (published 3 papers on PMB). We also proposed new methods for association analysis between patients (with esophageal cancer or schizophrenia) and healthy controls, with papers accepted by JTEMB and reviewed by BMJ Open.