Experimental design for nonparametric correction of misspecified dynamical models
From MaRDI portal
Publication:3176250
DOI10.1137/17M1128435zbMATH Open1403.62140arXiv1705.00956OpenAlexW2963466485MaRDI QIDQ3176250FDOQ3176250
Authors: Gal Shulkind, Lior Horesh, Haim Avron 
Publication date: 19 July 2018
Published in: SIAM/ASA Journal on Uncertainty Quantification (Search for Journal in Brave)
Abstract: We consider a class of misspecified dynamical models where the governing term is only approximately known. Under the assumption that observations of the system's evolution are accessible for various initial conditions, our goal is to infer a non-parametric correction to the misspecified driving term such as to faithfully represent the system dynamics and devise system evolution predictions for unobserved initial conditions. We model the unknown correction term as a Gaussian Process and analyze the problem of efficient experimental design to find an optimal correction term under constraints such as a limited experimental budget. We suggest a novel formulation for experimental design for this Gaussian Process and show that approximately optimal (up to a constant factor) designs may be efficiently derived by utilizing results from the literature on submodular optimization. Our numerical experiments exemplify the effectiveness of these techniques.
Full work available at URL: https://arxiv.org/abs/1705.00956
Recommendations
Nonparametric regression and quantile regression (62G08) Learning and adaptive systems in artificial intelligence (68T05) Gaussian processes (60G15) Optimal statistical designs (62K05) Simulation of dynamical systems (37M05)
Cites Work
- Bayesian experimental design: A review
- Bayesian calibration of computer models. (With discussion)
- A threshold of ln n for approximating set cover
- Gaussian processes for machine learning.
- Theory of Reproducing Kernels
- Kernels for vector-valued functions: a review
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Expected information as ecpected utility
- An arbitrary Lagrangian-Eulerian finite element method for transient dynamic fluid-structure interactions
- Nonlinear model reduction via discrete empirical interpolation
- Numerical initial value problems in ordinary differential equations.
- Title not available (Why is that?)
- Approximation of Large-Scale Dynamical Systems
- Lattice Boltzmann model for the incompressible Navier-Stokes equation
- Submodular functions and optimization.
- Title not available (Why is that?)
- Goal-oriented inference: approach, linear theory, and application to advection diffusion
- Dynamics of Bayesian updating with dependent data and misspecified models
- Randomized algorithms for estimating the trace of an implicit symmetric positive semi-definite matrix
- Universal kernels
- Introduction to Derivative-Free Optimization
- MultiScale Modeling of Physical Phenomena: Adaptive Control of Models
- Bounds for Truncation Error of the Sampling Expansion
- An analysis of approximations for maximizing submodular set functions—I
- Predicting the output from a complex computer code when fast approximations are available
- Computation of the Complex Error Function
- Title not available (Why is that?)
- Numerical Treatment of Defective Boundary Conditions for the Navier--Stokes Equations
- Analytical approximate solutions for a general nonlinear resistor-nonlinear capacitor circuit model
- Assessing the influence of heart rate in local hemodynamics through coupled 3D-1D-0D models
- Probabilistic model identification of uncertainties in computational models for dynamical systems and experimental validation
- A knowledge-based approach to response surface modelling in multifidelity optimization
- Maximizing a Submodular Set Function Subject to a Matroid Constraint (Extended Abstract)
- A Bayesian approach to model inadequacy for polynomial regression
- Dynamic data-driven reduced-order models
- Optimal value of information in graphical models
- Fast simulation of 3D electromagnetic problems using potentials.
- Flow rate defective boundary conditions in haemodynamics simulations
- Title not available (Why is that?)
- Numerical methods for the design of large-scale nonlinear discrete ill-posed inverse problems
- Numerical methods for experimental design of large-scale linear ill-posed inverse problems
- Statistical inference for dynamical systems: a review
- An extension of Mercer theorem to matrix-valued measurable kernels
- Faster Kernel Ridge Regression Using Sketching and Preconditioning
- Experimental design in the context of Tikhonov regularized inverse problems
- Nuclear Norm Optimization and Its Application to Observation Model Specification
Cited In (6)
- Optimal Design of Large-scale Bayesian Linear Inverse Problems Under Reducible Model Uncertainty: Good to Know What You Don't Know
- Optimal experimental design for infinite-dimensional Bayesian inverse problems governed by PDEs: a review
- Optimal design of large-scale nonlinear Bayesian inverse problems under model uncertainty
- Differential equations in data analysis
- Kernel-based parameter estimation of dynamical systems with unknown observation functions
- Data-Driven Discovery of Closure Models
This page was built for publication: Experimental design for nonparametric correction of misspecified dynamical models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3176250)
