Developed an autoencoder model to detect outliers in high-dimensional brain imaging data (MRI, PET, EEG), which utilized reconstruction error metrics for identifying statistical outliers, enhancing detection beyond visual assessments.
Fine-tuned a Segment Anything Model for brain tumor segmentation, and using Statistical Parametric Mapping to enhance the performance
ByteDance Ltd.
Big Data Engineer Internship
Built data ETL pipelines on content review traffic event logs, performed data wrangling and aggregation to help transport data from an operation data store (ODS) layer to a data warehouse for faster accessibility and better security
Implemented a mathematical programming solution with Python to solve for a stratified sampling sample size that satisfies a specific distribution, bringing the sample characteristics closer to reality and facilitating the business side to make decisions
Sichuan New Energy Vehicle (NEV) Innovation Center
Co-op Internship for Modelling Engineer
Cooperatively developed forecasting models based on macroeconomic indicators (top-down) and sectoral energy demand (bottom-up) to investigate the impact of NEV policy decision-making in Python to give the comprehensive assessment
Provided statistical and quantitative evaluation analysis, created vivid visualizations, and identified that policies of NEV changes and clean energy applications will have changes on greenhouse gas emissions and economic
Education
Univeristy of Copenhagen
M.Sc. in IT and Cognition
Renmin University of China
B.S. in Mathematics & B.A. in Economics
Minor in Cognitive Science
GPA: 3.7/4.0 (Rank: top 15%)
Graduation Thesis: Green Multiplier: A Dynamic Discrete Choice Model of NEV Market and Fiscal Policy Analysis in China
Proposed a dynamic discrete choice model to solve a nonlinear optimization problem for household car-buying decision-making in Python using vectorization and Numba library, greatly speeding up the high-dimensional computing efficiency from hours to seconds, the code can be found in my github.
Generated a novel indicator of carbon emission reduction to fiscal expenditure ratio based on a heterogeneous household model and conducted a counterfactual analysis to investigate the effects of fiscal policies in the new energy vehicle market on carbon emission reduction.
Awards: the 1st Prize of Excellent Graduation Thesis for Undergraduates of RUC (top 3%) and the phased achievements of the National Natural Science Foundation of China (approval number: 71973143); presented papers at the 8th China Public Finance Forum and the 8th Young & Middle-aged Fiscal Scholars Forum, 6th Applied Economics Advanced Frontier Forum of Chinese Industry Economy etc.