Machine Learning

Target-Aware Linear Regression Under Distribution Shift
Distribution shift between training and deployment is a pervasive challenge for modern AI systems. In many cases, the target marginals …
PDF Cite Code DOI arXiv GitHub (TZstats-Columbia)
Crowdsourcing the frontier: Advancing hybrid physics-ML climate simulation via a $50,000 Kaggle competition
PDF Cite DOI Journal
Structured exploration of machine learning model complexity for spatio-temporal forecasting of urban flooding
PDF Cite DOI Journal SSRN preprint
Power Ensemble Aggregation for Improved Extreme Event AI Prediction
This paper addresses the critical challenge of improving predictions of climate extreme events, specifically heat waves, using machine …
PDF Cite DOI arXiv
Probabilistic data assimilation for ensemble distribution projections with generative machine learning: A Lorenz '96 proof-of-concept
PDF Cite DOI Journal ESSOAr preprint
ClimSim-Online: A Large Multi-Scale Dataset and Framework for Hybrid Physics-ML Climate Emulation
PDF Cite Code Journal arXiv
Machine-assisted physical closure for coarse suspended sediments in vegetated turbulent channel flows
The parameterization of suspended sediments in vegetated flows presents a significant challenge, yet it is crucial across various …
PDF Cite DOI Journal
Spatiotemporal upscaling of sparse air-sea pCO2 data via physics-informed transfer learning
PDF Cite DOI Journal
Toward a Taxonomy of Trust for Probabilistic Machine Learning
Probabilistic machine learning increasingly informs critical decisions in medicine, economics, politics, and beyond. We need evidence …
PDF Cite DOI Journal arXiv
Next Waves in Veridical Network Embedding
Embedding nodes of a large network into a metric (e.g., Euclidean) space has become an area of active research in statistical machine …
Cite DOI arxiv