Research

The Human-Centered Machine Learning Lab explores how AI systems can be made more trustworthy, explainable, and beneficial — with an emphasis on privacy-sensitive domains like healthcare and education.

Focus areas

What my lab works on

Trustworthy & Private Machine Learning

Developing federated learning and differential privacy techniques that let sensitive data — especially in healthcare — be used for training without compromising individual privacy.

Human-Centered AI

Studying how people build trust and mental models of AI systems, and designing interaction techniques that support meaningful human oversight.

Explainable & Interpretable Systems

Creating visualization and analysis tools that make complex models — including graph neural networks and large language models — easier to understand and debug.

AI for Education

Building adaptive tutoring systems that respond to learners' cognitive load and support equitable access to personalized instruction.

Current work

Active & recent projects

Active · NIH-funded

FedCare: Privacy-Preserving Clinical Language Models

A multi-institution collaboration building federated NLP models for clinical decision support without centralizing patient data.

Active · NSF-funded

GlassBox: Visual Debugging for Graph Neural Networks

An open-source toolkit for interactively inspecting attention and message-passing behavior in GNNs.

Completed 2024

EquiTutor: Adaptive Learning at Scale

A large-scale study of multimodal tutoring interfaces deployed across three partner school districts.

Support

Funding

AgencyProgramAmountYears
National Science FoundationCAREER Award — Trustworthy Federated Learning$650,0002021–2026
National Institutes of HealthR01 — Privacy-Preserving Clinical NLP$2,100,0002022–2027
Industry Gift, TechCorp ResearchUnrestricted gift supporting graduate fellowships$300,0002023–2025

Prospective students

I welcome inquiries from prospective PhD and MS students interested in trustworthy and human-centered AI. Please review my recent publications before reaching out.

Contact the lab