Fei Ding Email & Phone Number

United States

• Led the development of a model framework to identify high-quality content, applying an advanced machine learningmodel to significantly enhance user engagement.• Spearheaded a novel neural network, Batchformer, to address feature distribution disparities across platforms,optimizing for seamless cross-platform application analytics.• Innovated the Microbiome Model project to refine content recommendations through sophisticated user behavioranalysis, significantly improving… Show more • Led the development of a model framework to identify high-quality content, applying an advanced machine learningmodel to significantly enhance user engagement.• Spearheaded a novel neural network, Batchformer, to address feature distribution disparities across platforms,optimizing for seamless cross-platform application analytics.• Innovated the Microbiome Model project to refine content recommendations through sophisticated user behavioranalysis, significantly improving engagement metrics.• Achieved a 91% accuracy rate in the account linking and transition system, facilitating user experience continuityacross platforms by applying state-of-the-art machine learning algorithms.• Led the development of an automated AIGC Video Interpretation and Integrated Creation system, enhancing videocontent quality and production efficiency using advanced AI techniques. Show less

College Station, Texas, United States

College Station, Texas, United States

Warp-U MCMC and stochastic bridge sampling: Developed a novel adaptive MCMC method for efficient Bayesian evidence estimation, addressing challenges in multi-modal target distributions. Demonstrated the method’s efficacy in a simulation study and an astronomical data analysis.• Functional PCA with covariate dependent mean and covariance structure: Introduced a new framework, CD-FPCA, enhancing predictive performance by incorporating covariates into functional PCA. Validated the approach… Show more Warp-U MCMC and stochastic bridge sampling: Developed a novel adaptive MCMC method for efficient Bayesian evidence estimation, addressing challenges in multi-modal target distributions. Demonstrated the method’s efficacy in a simulation study and an astronomical data analysis.• Functional PCA with covariate dependent mean and covariance structure: Introduced a new framework, CD-FPCA, enhancing predictive performance by incorporating covariates into functional PCA. Validated the approach through comprehensive simulation studies and astronomical data analysis. Show less

Doctor Of Philosophy - Phd, Statistics, 4.0

Advisor: Dr. David Jones.

Doctor Of Philosophy - Phd, Statistics, 3.81

Advisor: Dr. Shiyuan He.

Bachelor'S Degree, Finance, General

I study the second major in Finance and get a bachelor's degree from Wuhan University.

Bachelor'S Degree, Economics And Mathematics