Zhicheng (Stark) Guo
I’m a postdoctoral researcher at Duke University working with Professor Cynthia Rudin. I received my Ph.D. in Electrical and Computer Engineering from Duke in 2026 and my B.Sc. in Computer Science from Rensselaer Polytechnic Institute.
My research develops machine learning methods that turn the rich representations learned by foundation models into general, reusable, and human-understandable concepts for scientific discovery, prediction, and expert reasoning. This work lies at the intersection of scientific discovery, foundation models, and interpretable machine learning.
Currently, I am developing unsupervised concept-discovery methods and an interpretable lexicon of mammographic textures for breast cancer risk prediction. Across critical-care EEG and wearable cardiac monitoring, my work has provided the first data-driven support for the ictal-interictal-injury continuum (IIIC) and uncovered a shared manifold across ECG and PPG signals that improves atrial-fibrillation detection. My work has appeared in NEJM AI, Nature Machine Intelligence, JMLR, CVPR, and Harvard Data Science Review.
news
| Jul 01, 2026 | I started a new role as a postdoctoral researcher at Duke University, developing foundation models and interpretable AI for medical imaging. |
|---|---|
| May 01, 2026 | I completed my Ph.D. in Electrical and Computer Engineering at Duke University! |
| Aug 30, 2025 | Wrapped up my machine learning research internship at Texas Instruments Kilby Labs! |
| May 30, 2025 | I will attend CVPR 2025 at Nashville. See you there! |
| May 14, 2025 | Started my summer internship as a Machine Learning Researcher at Kilby Labs @ Texas Instrument, Dallas, TX. I will be working on LLM and Embed AI. |
selected publications
-
ProtoCDisco: Unsupervised Concept Discovery Using Foundation Model RepresentationsManuscript in submission, 2026 -
"What is Different Between These Datasets?" A Framework for Explaining Data Distribution ShiftsJournal of Machine Learning Research, 2025 -
Rashomon sets for prototypical-part networks: Editing interpretable models in real-timeIn Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Jun 2025 -
Improving Clinician Performance in Classifying EEG Patterns on the Ictal–Interictal Injury Continuum Using Interpretable Machine LearningNEJM AI, Jun 2024 -
Improving Atrial Fibrillation Detection Using a Shared Latent Space for ECG and PPG SignalsHarvard Data Science Review, Jun 2025 -
Learning From Alarms: A Robust Learning Approach for Accurate Photoplethysmography-Based Atrial Fibrillation Detection using Eight Million Samples Labeled with Imprecise Arrhythmia AlarmsIEEE Journal of Biomedical and Health Informatics, Jun 2024 -
Sparse learned kernels for interpretable and efficient medical time series processingNature Machine Intelligence, Jun 2024 -
A supervised machine learning semantic segmentation approach for detecting artifacts in plethysmography signals from wearablesPhysiological Measurement, Jun 2021
