Yijin Ni (倪亦瑾)

I am a Ph.D. candidate under the Statistics track of the Industrial and Systems Engineering school at Georgia Institute of Technology, fortunate to be advised by Prof. Xiaoming Huo.

Prior to Georgia Tech, I received my Bachelor’s degree in Statistics in 2020 from the University of Science and Technology of China (USTC), advised by Prof. Canhong Wen.

I am broadly in the utilization of statistical tools in the exploration of methodologies. Specifically, my research topics include:

• Nonlinear uniform concentration inequality, regarding theoretical boundaries for optimization problems;
• Kernel-based statistics, including Maximum Mean Discrepancy (MMD), Hilbert-Schmidt Independence Criterion (HSIC), Energy Distance (ED), and distance Covariance (dCov);
• Dimension reduction and variable selection, serving as a preprocessing step for high dimensional statistics;
• Fairness representation learning, exploring available metrics to achieve fairness for a wide range of downstream tasks;
• Preference learning for LLM fine-tuning, including Direct Prefernece Optimization (DPO) and Reinforcement Learning from Human Feedback (RLHF).