Preprint

A framework for assessing and understanding sources of error in Earth System Model emulation
PDF Cite DOI
Reduced-Order Modeling of Cyclo-Stationary Time Series Using Score-Based Generative Methods
PDF Cite DOI
Score-based generative emulation of impact-relevant Earth system model outputs
PDF Cite DOI
Statistical Parameter Calibration with the Generalized Fluctuation Dissipation Theorem and Generative Modeling
PDF Cite DOI
Surface to Seafloor: A Generative AI Framework for Decoding the Ocean Interior State
This preprint presents a generative AI framework for inferring the interior state of the ocean from surface observations. The method …
PDF Cite DOI
Predicting Forced Responses of Probability Distributions via the Fluctuation-Dissipation Theorem and Generative Modeling
We present a flexible data-driven framework for estimating the response of higher-order moments of nonlinear stochastic systems to …
PDF Cite DOI
KGMM: A K-means Clustering Approach to Gaussian Mixture Modeling for Score Function Estimation
We propose a hybrid method for accurately estimating the score function – the gradient of the logarithm of a system’s …
PDF Cite DOI
Learning Dissipation and Instability Fields from Chaotic Dynamics
This paper establishes a novel theoretical connection between the operator-theoretic description of a dynamical system and its local …
PDF Cite DOI
A Modified Bisecting K-Means for Approximating Transfer Operators: Application to the Lorenz Equations
We investigate the convergence behavior of the extended dynamic mode decomposition for constructing a discretization of the continuity …
PDF Cite DOI
A Statistical Emulator Design for Averaged Climate Fields
This preprint introduces a statistical emulator designed to represent averaged climate fields. The emulator aims to efficiently …
PDF Cite DOI