AI for Science
Combining machine learning, scientific knowledge, and simulation to discover mechanisms and understand complex systems.
Scientific inference for discovery and decision-making.
We develop machine learning methods for scientific discovery and informed decisions, grounded in continuous-time modeling and evidence-centered evaluation.
Research directions
Continuous-Time Modeling
Learning continuous dynamics from irregular observations through neural differential equations, with attention to stability and uncertainty.
Evidence-Centered AI
Developing evaluation principles that align AI claims with robust comparisons, uncertainty assessment, and reproducible evidence.
Industrial AI & Decision-Making
Connecting machine learning, optimization, and simulation to support decisions in manufacturing, logistics, and service systems.
Latest news
Scientific Inference Lab opens
The Scientific Inference Lab opens at Pusan National University.
Silent failures at KDD 2026
The paper on silent failures in federated personalization appears in the KDD 2026 proceedings, Blue Sky Ideas Track.
Evidence standards at ICML 2026
A sole-authored position paper on evidence standards for machine learning was presented at ICML 2026.
Selected for the NVIDIA Academic Grant Program
The Neural SDE-Augmented GPU-Accelerated Pharmacokinetic Simulation Framework project, with Alex Bui at UCLA Medical Imaging Informatics, was selected for the NVIDIA Academic Grant Program.
Research opportunities & collaboration
We welcome inquiries about research opportunities and collaboration in machine learning and scientific inference.