· Winston Chen

About

I am a PhD candidate in the Computer Science & Engineering at the University of Michigan, advised by Prof. Jenna Wiens. I am interested in enabling reliable and interpretable AI-driven decision-making in healthcare, leveraging tools from causal inference and explainable AI.

Previously, I received my B.S. in Electrical Engineeirng at the University of Washington, during which I worked with Prof. William Stafford Noble and Prof. Yang Lu on biological interaction discovery with explainable AI.

Conditional Front-door Adjustment for Heterogeneous Treatment Assignment Effect Estimation Under Non-adherence
- Winston Chen, Trenton Chang, Jenna Wiens
- Conference on Health, Inference, and Learning (CHIL), 2025 [Paper] [Code]
Error-controlled Non-additive Interaction Discovery in Machine Learning Models
- Winston Chen, Yifan Jiang, William Stafford Noble, Yang Young Lu
- Nature Machine Intelligence, 2025 [Paper] [Poster] [Code]
- Previously presented at NeurIPS 2024 workshop on Interpretable AI Past, Present and Future and Machine Learning in Computational Biology (MLCB), 2023

Program Sub-Chair, Machine Learning for Health Symposim (ML4H), (2024 - present)
Reviewer, International Conference on Artificial Intelligence and Statistics (AISTATS), (2024)
Secretary, CSE Graduate Student Organization, University of Michigan, (2024 - present)

Mary Gates Research Scholarship, Mary Gates Endowment For Students, UW (2021)
- Scholarship supporting outstanding undergraduate research
Lawrence & Lucille Frey Endowed Electrical & Computer Engineering Scholarship, Department of Electrical & Computer Engineering, UW (2020)
- Scholarship awarded to ECE students with academic excellence
Herschel & Caryl Roman Scholarship, Department of Genome Science, UW (2020)
- Scholarship awarded to undergraduate students working in genetics research.

Built a FIDO2 roaming authenticator prototype that allows passwordless authentication on web services supporting WebAuthn.
Integrated the authenticator with SAFR facial recognition technology to enable biometric verification.
Demonstrated the authenticator in collaboration with StrongKey at the flagship FIDO conference, Authenticator 2022.

Built and deployed a testing infrastructure for analyzing and visualizing NVIDIA's genomics computing software logs.
Drove discussions with core developers regarding testing infrastructure's feature requirements.
Presented the final project results through a slide deck, confluence page, and detailed README.

Facilitated moving 500+ residents into Elm Resident Hall in the course of four days.
Advised Elm Hall council in organizing 20+ building-wide events during the 2022-2023 academic year.
Assisted the Elm Hall resident director in managing 10 resident advisors through performing administrative tasks such as scheduling on-calls.

Attended weekly meetings to provide updates on current corporate relationship projects.
Collaborated with industry recruiters (e.g. TI, Tesla, Wyze) to organize 10+ industry networking events.
Maintained positive relationship with industry sponsors through emails and check-in meetings.

Collaboratively planned and executed 30+ events every year to engage 150+ residents.
Created and maintained floor decorations to form a welcoming residential environment.
Regularly being on-call 5PM-8AM to provide residents with emergency assistance and secure the safety of the entire residence hall.

KiwiLink is an Android/iOS mobile app for online study buddy finding in college setting.
Rapidly built the prorotype app using JavaScript, React Native, and Firebase (Summer 2020).
Led a team of 10 engineers maintaining app operation and developing new features to serve 1500+ users at the University of Washington. (Sept 2020 - Oct. 2021)

Fasr is a portable facemask sanitizer device that allows user to efficiently clean their mask using heat and disinfectant.
Led a team of 6 people developed the hardware prototype and a business plan.
Participated in the UW Dempsey Startup Competition (2021).
Open sourced all the technical and business documentations.