New journal publications in 2026
Our work on tensor and sparse topic models and canonical correlation analysis is appearing in JMLR and the Electronic Journal of Statistics.
News & Announcements
Latest publicationsNew preprints from the group
Explore our latest work on data selection for fine-tuning, GNNs for brain connectomes, graph-based learning, network dynamics, and microbial communities.
SpeedCP at ICML 2026
Our students presented the group's work on fast, kernel-based conditional conformal prediction at ICML.
Prospective applicants
The lab is not currently accepting applications for student, research assistant, or postdoctoral positions.
What we do
See all projectsGNN Theory & Methods
We study statistical properties of GNNs to try and design interpretable, reliable graph learning pipelines.
Structured Estimation
We develop high‑dimensional estimators that leverage sparsity and known structure for robust inference.
Spatial Transcriptomics
We deploy our statistical frameworks to uncover spatially organized gene expression patterns and cell–cell interactions from high-resolution transcriptomic data.
Microbial GWAS
We use high-dimensional statistical methods to identify genetic and metabolic determinants of key microbial traits, such as thermotolerance and photosynthetic efficiency.
Past Events
Our students at ICML 2026
Sharing SpeedCP, our work on fast, kernel-based conditional conformal prediction, at the poster session.
SIGNAL Lab at NeurIPS 2025
Sowon, Yating, Yeo Jin, and Claire attended NeurIPS 2025 in San Diego.
Welcome to our newest lab members
The lab welcomes Maria, Coralie, Jeff, and Thomas, who will be working on biological applications.
Yeo Jin presented at the RSS meeting
She showcased the group's work on conformal prediction.
Sowon attended the MLoG-GenAI workshop at KDD
She presented her work on unsupervised learning for GNNs.
The group hosted a workshop at the NITMB!
The workshop highlighted recent advances and new perspectives in GNN theory.
Our group attended the SNAB workshop in Nassau!
We showcased our work on topic models, GNNs and CCA during the poster session.
The group hosted a workshop at IMSI!
We were lucky to have great speakers and learned a lot about challenges in the deployment of GNNs in the Physical Sciences.