Zonghan Zhang 张宗瀚
Ph.D. in Computer ScienceDepartment of Computer Science and Engineering
Mississippi State University, USA.
Email: zz239 [AATTT] msstate [DOT] edu
Office: Room 306, Butler Hall
CV | Research Statement | Teaching Statement | GitHub | Google Scholar | LinkedIn
Short Biography I received my Ph.D. in Computer Science from Mississippi State University in 2025, advised by Prof. Zhiqian Chen. My research focuses on uncertainty-aware and interpretable machine learning for complex networked systems, with an emphasis on physically informed modeling and decision-making under uncertainty.
A central theme of my work is to move beyond purely data-driven approaches by integrating structured inductive biases, probabilistic reasoning, and mechanistic insights into graph-based AI models.
Before my Ph.D., I received an M.S. in Information Science from the University of Pittsburgh.
News
2025. Our paper Sobol’s Total Indices for Accurate and Scalable Feature Exclusion in High-Dimensional Data is accepted by IEEE BigData 2025.
2025. Our paper Graph Symbolic Regression to Interpret the Spread of Vesicular Stomatitis Virus across the U.S. and Mexico is accepted by ACM SIGSPATIAL 2025.
Research
My research focuses on uncertainty-aware and interpretable machine learning for complex networked systems. I develop principled models to understand, predict, and control propagation processes under uncertainty, with applications in socio-technical systems such as information diffusion, epidemics, and infrastructure networks.
- Physically-informed and structured models for network flow and diffusion dynamics
- Uncertainty quantification and sensitivity analysis (e.g., Sobol indices)
- Probabilistic graph learning and Bayesian optimization for decision-making
- Interconnected networks and tensor representations for cross-layer systems
Publications
- Zonghan Zhang, Zijian Zhang, and Zhiqian Chen. Multiple-Source Localization from a Single-Snapshot Observation Using Graph Bayesian Optimization. (AAAI 2024) [ PDF, Code, Slides, Poster ]
- Zonghan Zhang, Zhiqian Chen. Understanding Influence Maximization via Higher-Order Decomposition. (SDM 2023) [ PDF, Code, Slides, Poster ]
- Zonghan Zhang, Zhiqian Chen. Sobol’s Total Indices for Accurate and Scalable Feature Exclusion in High-Dimensional Data. (IEEE BigData 2025) [ PDF, Slides ]
- Tamanna Rashme, Zonghan Zhang, Jason Weeks, Marouane Benbrahim, Zijian Zhang, Zhiqian Chen, Nisha Pillai, Ram Ramkumar, and Bindu Nanduri. Graph symbolic regression to interpret the spread of Vesicular Stomatitis Virus across the U.S. and Mexico. (ACM SIGSPATIAL 2025) [PDF]
- Zonghan Zhang, Ramyasri Veerapaneni, Moses Ayoola, Athish Ram Das, Zhiqian Chen, Bindu Nanduri, and Mahalingam Ramkumar. Leveraging Graph Neural Networks for MIC Prediction in Antimicrobial Resistance Studies. (IEEE EMBC 2024) [PDF]
- Zonghan Zhang, Zhiqian Chen. Physically-Informed SIR Model for Network Diffusion. (NetSci 2026 Abstract) [PDF]
- Zonghan Zhang, Zhiqian Chen. Directional Tensor Embeddings for Interconnected Network Propagation. (NetSci 2026 Abstract) [PDF]
- Zonghan Zhang, Subhodip Biswas, Fanglan Chen, Kaiqun Fu, Taoran Ji, Chang-Tien Lu, Naren Ramakrishnan, and Zhiqian Chen. Blocking Influence at Collective Level with Hard Constraints. (AAAI 2022 Student Abstract) [PDF, CODE]
Research Systems
- Co-Simulation of Power and Transportation Networks (SUMO + PyPSA). A coupled simulation framework for modeling interactions between transportation and power systems, including EV charging, infrastructure dependencies, and cascading failures. [PDF]
Teaching & Mentorship
My teaching philosophy emphasizes helping students understand not only how to build machine learning models, but how to reason about their reliability, limitations, and broader impact. I focus on developing structured thinking, uncertainty awareness, and interpretability alongside technical skills.
-
Guest Lecturer & Teaching Assistant, Mississippi State University
Courses: Machine Learning, Artificial Intelligence, Graph Machine Learning
I have mentored students on research projects in graph learning, epidemic simulation, and interpretable modeling, guiding them from problem formulation to experimental evaluation and analysis.
I am prepared to teach courses including Data Structures and Algorithms, Machine Learning, Graph Machine Learning, Network Science, Artificial Intelligence, and Data Mining.
For more details, please see my Teaching Statement.