Bayesian statistics

Bayesian Inference

Sampling

MCMC

Theoretical Machine Learning

Deep Learning

Interest Keywords

I’m drawn to the Bayesian approach to thinking about uncertainty, so its application to biomedical data is of interest to me. I am interested in the MCMC and Variational Inference for large data sets, the use of Bayesian methods in deep learning, the semi-parametric Bayesian. I have an ongoing project integrating Bayesian with RNN in simultaneously handles MNAR/MAR missingness and left-censoring in Alzheimer’s longitudinal data. I have a post project extending a cognitive measurement method under the Bayesian framework to handle left-censored data.

Citation

BibTeX citation:

@online{untitled,
  author = {},
  title = {Bayesian Statistics},
  url = {https://kaizhongmu.github.io/research/MNAR/},
  langid = {en}
}

For attribution, please cite this work as:

“Bayesian Statistics.” n.d. https://kaizhongmu.github.io/research/MNAR/.

---
title: "Bayesian statistics" 
categories: [Bayesian Inference, Sampling, MCMC, Theoretical Machine Learning, Deep Learning]
author: ""
date: ""
image: 1.png
citation: true
---  

## Interest Keywords

I'm drawn to the Bayesian approach to thinking about uncertainty, so its application to biomedical data is of interest to me. I am interested in the MCMC and Variational Inference for large data sets, the use of Bayesian methods in deep learning, the semi-parametric Bayesian. I have an ongoing project integrating Bayesian with RNN in simultaneously handles MNAR/MAR missingness and left-censoring in Alzheimer's longitudinal data. I have a post project extending a cognitive measurement method under the Bayesian framework to handle left-censored data.

## Related Experience

- **Research Project**: A Bayesian Cognitive Change Method for Censored Longitudinal Data
- **Research Project**: An Bayesian-based RNN Method for MNAR/MAR + left-censored
- **Manuscript**: A Reliable Change Estimation Method for Severely Cognitively Impaired Populations

## Related Story

The story of Bayesian statistics and me began with a research project where I encountered a non-identifiability issue when using MLE to fit a mixed-effects model. Learning that Bayesian inference might offer a solution, I enrolled in a Bayesian course and received systematic training—from explicit likelihood to MCMC methods for non-explicit likelihoods. As the course concluded, another project of mine entered the stage of method development. The Bayesian framework inspired me to address left-censored data with a simpler mathematical formulation while providing uncertainty quantification for clinicians. Building on this foundation, I am now conducting Bayesian inference under an RNN architecture to handle MNAR and left-censored data simultaneously.