Machine learning researcher focused on building models for biological sequence data.
I am a PhD candidate in the Joint Carnegie Mellon University- University of Pittsburgh PhD Program in Computational Biology, where I work on AI/ML research for immunology.
My research interests include statistical modeling and deep learning for noisy, high-dimentionsal signals.
Currently, I am developing an interpretable variant of the attention mechanisms for transformer models. Highlighted Publications and Projects (scroll down for more):
Interpretable machine learning uncovers epithelial transcriptional rewiring and a role for Gelsolin in COPD
J Su *, H Xiao *, et al JCI Insight, * represents equal contribution Link to Manuscript
From bench to bedside via bytes: Multi-omic immunoprofiling and integration using machine learning and network approaches H Xiao , et al Human Vaccines & ImmunotherapeuticsLink to Manuscript
Spatial microniches of IL-2 combine with IL-10 to drive lung migratory TH2 cells in response to inhaled allergen
k He, H Xiao , et al Nature ImmunologyLink to Manuscript
September 2024: I recieved research presentation award for my project titled 'Interpretable machine learning uncovers spatial-microenvironment-specific drivers of pathogenesis in myocardial infarction' at the University of Pittsburgh Immunology Retreat.
May 2024: My co-first-authored paper where we developed a novel interpretable machine learning model, SLIDE, has been published at Nature Methods.
December 2023: I have been awarded the Quantitative Methodologies Pilot Program (QuMP) grant as a co-principle investigator for the project "Uncovering immunomodulatory spatial microniches in asthma involving Th2 and Tfh2 cells using interpretable machine learning ."