About me

I am a third-year biostatistic major Ph.D. student at the University of Florida and I am fortunate to work with Prof. Qing Lu at Statistical Learning Lab. My research interests lie primarily in statistical learning models and its application on the counting processes data (mainly the survival data) analysis.

I got my M.S. from Michigan State University, where I was getting into machine learning and started to do some exploratory research under the guidance of Prof. Xiaoming Liu. I got my B.S. from Tongji University, where I strengthened my foundation of mathematics.

Projects

Kernel-Based Neural Network (KNN) for Cox Proportional Hazard Function with High-dimensional Genetic Data

  • Discovered traditional cox model does not perform well in nonlinear underlying function or high- dimensional data
  • Designed and implemented a kernel-based neural network to estimate the variance components in linear mixed model
  • Demonstrated that the KNN model outperforms competitors like traditional Cox model with elastic-net regularization and PyCox model by 18% in terms of C-index using DLBCL dataset

Neural-Network Transformation Models for Counting Processes

  • A neural network model to predict both the baseline cumulative hazard function and mis-specifying underlying function in the counting processes
  • Demonstrated that the neural-network transformation model outperforms the linear transforma- tion model by 16% in terms of estimation and 11% in terms of prediction accuracy when the covariate effects are nonlinear

Prediction and Hypothesis Testing for Interval Censored Competing risks Data via Kernel Machine Semiparametric Transformation Models

  • A kernel machine semiparametric transformation model to do risk prediction and association tests for interval censored competing risks data
  • Illustrated the proposed method is more robust than other testing methods and has more accurate predictions

Personal

I love everthing about nature.
I also play badminton and tennis in free time.
I have two doodle dogs, Noopy and Hugsy. They are cute girls.