CV

Ph.D.

Texas A&M University, United States

Information Science and Learning Systems

• Advisor: Prof. Chao Tian
• GPA: 4.0/4.0

M.S.

University of California San Diego, United States

Communication Theory and Systems

• Advisor: Prof. Xinyu Zhang

B.E.

Southeast University, China

Information Engineering

• Advisor: Prof. Cheng Zhang

Departmental Poster Event Winner

Texas A&M University

Sun Qingyun Scholarship

Southeast University

First Place in Mathematical Contest in Modeling (CUMCM)

China Society for Industrial and Applied Mathematics

Top 0.7%, best from SEU

Mitsubishi Electric Corporation Scholarship

Southeast University

Optimizing Leaky Private Information Retrieval Codes to Achieve O(log K) Leakage Ratio Exponent

Accepted by ISIT 2025

From Deep Additive Kernel Learning to Last-layer Bayesian Neural Networks via Induced Prior Approximation

Presented at AISTATS 2025

Acceptance rate: 31.3%.

Weakly Private Information Retrieval from Heterogeneously Trusted Servers

Submitted to IEEE TIT

Preprint. Under review.

Deep Gaussian Processes

A sparse expansion for deep Gaussian processes (DGPs) as Bayesian neural networks.

• Deep Additive Kernel Learning
• DMGP Package

Coding for Privacy

Optimizing code schemes for private information retrieval (PIR) systems.

• Weakly Private Information Retrieval
• Leaky Private Information Retrieval

Dynamic mmWave Networking

Machine learning can help automate network control decisions, but its practical deployment faces new system-level challenges in 5G IAB, from heterogeneous network slices to run-time decision-making.

• Self-supervised Adaptive Configuration
• Reinforcement Learning Mesh Controller