Ziping Zhao

My research interests lie at the intersection of machine learning, signal processing, optimization, applied probability, and statistics, with applications to emerging problems in data science and AI. Our group develops mathematical and computational methods for learning, inference, sensing, and decision-making in complex and networked systems. The overarching objective of our research is to bring theoretically grounded and computationally efficient methods together with emerging data science problems, so that methodological advances and practical applications can inform and reinforce each other. In pursuit of these objectives, the research in our group is organized into the following topics.

Optimization and Statistical Learning for Complex and High-Dimensional Data

Modern datasets are increasingly high-dimensional, heterogeneous, imperfect, and distributed across multiple locations. These characteristics pose fundamental challenges to statistical reliability and computational efficiency, particularly when conventional modeling assumptions are no longer appropriate. A central objective is to develop and analyze optimization and statistical learning methods under structural, latent-variable, and diffusion-based probabilistic models, accommodating complex data structures and realistic uncertainties while retaining meaningful statistical and computational properties.

Graph Neural Networks and Signal Processing

Many forms of data are naturally supported on graphs, where relationships among entities provide information beyond individual observations. Graph neural networks and graph signal processing offer complementary approaches to representing, analyzing, and learning from such relational data. Of particular interest are the connections between model-based signal processing and data-driven learning, as well as the roles of graph structure in representation, stability, robustness, interpretability, and generalization.

Distributed Machine Learning and Signal Processing over Networks

In many systems, data, information, and computational resources are distributed across multiple interconnected agents. Learning and signal processing in these settings require agents to cooperate through limited local communication without relying on centralized data collection. We investigate how learning, inference, optimization, and information exchange interact, seeking decentralized methods that balance statistical accuracy, computational cost, communication efficiency, robustness, and scalability.

Structured Inverse Problems for Information Acquisition and Sensing

Information acquisition often requires recovering unknown signals, matrices, or physical quantities from measurements that are indirect, incomplete, noisy, or nonlinear. Representative examples include blind deconvolution, blind calibration, phase retrieval, and low-rank recovery, which arise in imaging, sensing, data analysis, and other scientific and engineering applications. We investigate nonconvex optimization and statistical inference under structural and diffusion-based probabilistic models, with the goal of developing scalable recovery methods and establishing theoretical guarantees for identifiability, sample complexity, recovery accuracy, and algorithmic convergence.

Optimization and Decision-Making in Communication, Radar, and Financial Systems

Communication, radar, and financial systems give rise to structured optimization problems involving the design and allocation of limited resources under uncertainty. Our interests include beamforming design for wireless communications, waveform design for radar and sensing, and portfolio design for financial investment and risk management. Although these problems have different physical and economic interpretations, they share common structures in modeling, optimization, and learning. Motivated by these connections, we seek principled and computationally practical methods that respect domain-specific constraints and enable efficient, interpretable, and adaptive system design and decision-making.