Sue Zheng

- Developed an iterative approach to sequential experimental design that achieves significant computational savings through cost-aware allocation of resources. Presented results to peers at Neurips 2020 and sponsors at ETI 2020 and UPR 2021 Paper: Sue Zheng, David S. Hayden, Jason Pacheco, John W. Fisher III, “Sequential Bayesian Experimental Design with Variable Cost Structure,” in Neural Information Processing Systems, 2020 - Developed an approach to maximally reuse computation during information-theoretic planning. Presented results to peers at ICML 2018 and sponsors at UPR 2019 Paper: Sue Zheng, Jason Pacheco, John W. Fisher III, “A Robust Approach to Sequential Information Theoretic Planning,” in International Conference on Machine Learning, 2018 - Developed a driver behavior model used to predict a driver’s route and destination Paper: Julian Straub, Sue Zheng, John W. Fisher III, “Bayesian Nonparametric Modeling of Driver Behavior,” in IEEE Intelligent Vehicles Symposium, 2014 - Mentored junior PhD candidates and frequently reviewed colleagues’ papers - Contributed to on-going quarterly reports to sponsors and presented updates at annual reviews

- Created algorithms and performed analysis for many projects involving tracking and fusion, estimation, machine learning, and radar processing - Led the development of algorithms and software for the data fusion component of a large system, resulting in a successful demonstration - Developed a method to refine content-based video retrieval results using feedback that yielded 5% increase in probability of detection and 20% decrease in false alarms and that contributed to follow-on programs - Developed prediction models for tracking that exploit road context to describe typical vehicle behaviors, such as turning and passing - Provided mentorship to junior staff working on performance enhancement of tracking and fusion algorithms - Co-inventor on a patent application for a low-shot learning approach

- Performed research and experimentation culminating in Master’s thesis - Explored estimation algorithms for use in the tracking system of a free-space optical link - Simulated an airborne communications link, including the tracking system (consisting of the tracking servo and fast-steering mirror) and the impact of the boundary layer arising from the shape of the turret - Compared the performances of estimation algorithms using the communications link simulation