Machine learning moment closure models for the radiative transfer equation II: enforcing global hyperbolicity in gradient based closures

• URL: http://arxiv.org/abs/2105.14410v1
• Date: Sun, 30 May 2021 02:10:59 GMT
• Title: Machine learning moment closure models for the radiative transfer equation II: enforcing global hyperbolicity in gradient based closures
• Abstract summary: This is the second paper in a series in which we develop machine learning (ML) moment closure models for the radiative transfer equation (RTE) We propose a method to enforce the global hyperbolicity of the ML closure model. It is shown that the new ML closure system inherits the dissipativeness of the RTE and preserves the correct diffusion limit as the Knunsden number goes to zero.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: This is the second paper in a series in which we develop machine learning (ML) moment closure models for the radiative transfer equation (RTE). In our previous work \cite{huang2021gradient}, we proposed an approach to directly learn the gradient of the unclosed high order moment, which performs much better than learning the moment itself and the conventional $P_N$ closure. However, the ML moment closure model in \cite{huang2021gradient} is not able to guarantee hyperbolicity and long time stability. We propose in this paper a method to enforce the global hyperbolicity of the ML closure model. The main idea is to seek a symmetrizer (a symmetric positive definite matrix) for the closure system, and derive constraints such that the system is globally symmetrizable hyperbolic. It is shown that the new ML closure system inherits the dissipativeness of the RTE and preserves the correct diffusion limit as the Knunsden number goes to zero. Several benchmark tests including the Gaussian source problem and the two-material problem show the good accuracy, long time stability and generalizability of our globally hyperbolic ML closure model.

Related papers

• A Deep Second-Order Stochastic Residual Method for Fully Nonlinear Parabolic PDEs [0.9023847175654603]
  We introduce the Deep Second-Order Residual Method (D2SRM) for high-dimensional, Hessian-dependent fully nonlinear parabolic PDEs.<n>A single space-time network generates derivative-consistent approximations of the solution, gradient, and Hessian.<n>Experiments on a 100-dimensional manufactured benchmark compare terminal treatments, probe Hessian couplings inside and outside the proved small-gain range, and show decreasing errors as the time step decreases.
  arXiv  Detail & Related papers  (2026-07-18T09:43:15Z)
• A Hyperbolic Neural Closure for M1 Radiation Transfer [9.065972458046089]
  In radiation transfer simulations, an M1 method achieves substantial computational savings by replacing the full angular transport equation with a low-order moment system.<n>Because this reduced system is not closed, a closure model is required to represent the unknown higher-order moments.<n>We propose a hyperbolic neural closure for the M1 radiative transfer system to guarantee real eigenvalues.
  arXiv  Detail & Related papers  (2026-07-11T15:43:00Z)
• Regularized Online RLHF with Generalized Bilinear Preferences [68.44113000390544]
  We consider the problem of contextual online RLHF with general preferences.<n>We adopt the Generalized Bilinear Preference Model to capture preferences via low-rank, skew-symmetric matrices.<n>We prove that the dual gap of the greedy policy is bounded by the square of the estimation error.
  arXiv  Detail & Related papers  (2026-02-26T15:27:53Z)
• Learning quadratic neural networks in high dimensions: SGD dynamics and scaling laws [21.18373933718468]
  We study the optimization and sample complexity of gradient-based training of a two-layer neural network with quadratic activation function in the high-dimensional regime.<n>We present a sharp analysis of the dynamics in the feature learning regime, for both the population limit and the finite-sample discretization.
  arXiv  Detail & Related papers  (2025-08-05T17:57:56Z)
• Global Convergence of Gradient EM for Over-Parameterized Gaussian Mixtures [53.51230405648361]
  We study the dynamics of gradient EM and employ tensor decomposition to characterize the geometric landscape of the likelihood loss.<n>This is the first global convergence and recovery result for EM or gradient EM beyond the special case of $m=2$.
  arXiv  Detail & Related papers  (2025-06-06T23:32:38Z)
• Hyperbolic Machine Learning Moment Closures for the BGK Equations [0.0]
  We introduce a hyperbolic closure for the Grad moment expansion of the Bhatnagar-Gross-Krook's kinetic model using a neural network (NN) trained on BGK's moment data. This closure is motivated by the exact closure for the free streaming limit that we derived in our paper on closures in transport citeHuang2022-RTE1.
  arXiv  Detail & Related papers  (2024-01-09T19:14:57Z)
• Neural Inference of Gaussian Processes for Time Series Data of Quasars [72.79083473275742]
  We introduce a new model that enables it to describe quasar spectra completely. We also introduce a new method of inference of Gaussian process parameters, which we call $textitNeural Inference$. The combination of both the CDRW model and Neural Inference significantly outperforms the baseline DRW and MLE.
  arXiv  Detail & Related papers  (2022-11-17T13:01:26Z)
• Gradient-Free Methods for Deterministic and Stochastic Nonsmooth Nonconvex Optimization [94.19177623349947]
  Non-smooth non optimization problems emerge in machine learning and business making. Two core challenges impede the development of efficient methods with finitetime convergence guarantee. Two-phase versions of GFM and SGFM are also proposed and proven to achieve improved large-deviation results.
  arXiv  Detail & Related papers  (2022-09-12T06:53:24Z)
• $\texttt{FedBC}$: Calibrating Global and Local Models via Federated Learning Beyond Consensus [66.62731854746856]
  In federated learning (FL), the objective of collaboratively learning a global model through aggregation of model updates across devices tends to oppose the goal of personalization via local information. In this work, we calibrate this tradeoff in a quantitative manner through a multi-criterion-based optimization. We demonstrate that $texttFedBC$ balances the global and local model test accuracy metrics across a suite datasets.
  arXiv  Detail & Related papers  (2022-06-22T02:42:04Z)
• High-dimensional limit theorems for SGD: Effective dynamics and critical scaling [6.950316788263433]
  We prove limit theorems for the trajectories of summary statistics of gradient descent (SGD) We show a critical scaling regime for the step-size, below which the effective ballistic dynamics matches gradient flow for the population loss. About the fixed points of this effective dynamics, the corresponding diffusive limits can be quite complex and even degenerate.
  arXiv  Detail & Related papers  (2022-06-08T17:42:18Z)
• Machine learning moment closure models for the radiative transfer equation III: enforcing hyperbolicity and physical characteristic speeds [0.0]
  Third paper in a series in which we develop machine learning (ML) moment closure models for the radiative transfer equation (RTE) Motivated by the observation that the coefficient matrix of the closure system is lower Hessenberg matrix, we relate its eigenvalues to the roots of an associated neural network.
  arXiv  Detail & Related papers  (2021-09-02T04:16:56Z)
• Machine learning moment closure models for the radiative transfer equation I: directly learning a gradient based closure [0.0]
  We propose to learn the gradient of the high order moment using neural networks. A variety of benchmark tests, including the variable scattering problem, show both good accuracy and generalizability of our machine learning closure model.
  arXiv  Detail & Related papers  (2021-05-12T14:24:20Z)
• Last iterate convergence of SGD for Least-Squares in the Interpolation regime [19.05750582096579]
  We study the noiseless model in the fundamental least-squares setup. We assume that an optimum predictor fits perfectly inputs and outputs $langle theta_*, phi(X) rangle = Y$, where $phi(X)$ stands for a possibly infinite dimensional non-linear feature map.
  arXiv  Detail & Related papers  (2021-02-05T14:02:20Z)
• Stochastic Hamiltonian Gradient Methods for Smooth Games [51.47367707388402]
  We focus on the class of Hamiltonian methods and provide the first convergence guarantees for certain classes of smooth games. Using tools from the optimization literature we show that SHGD converges linearly to the neighbourhood of a gradient. Our results provide the first global non-asymotic last-rate convergence guarantees for the class of general games.
  arXiv  Detail & Related papers  (2020-07-08T15:42:13Z)
• Differentiating through the Fr\'echet Mean [51.32291896926807]
  Fr'echet mean is a generalization of the Euclidean mean. We show how to differentiate through the Fr'echet mean for arbitrary Riemannian manifold. This fully integrates the Fr'echet mean into the hyperbolic neural network pipeline.
  arXiv  Detail & Related papers  (2020-02-29T19:49:38Z)
• A Near-Optimal Gradient Flow for Learning Neural Energy-Based Models [93.24030378630175]
  We propose a novel numerical scheme to optimize the gradient flows for learning energy-based models (EBMs) We derive a second-order Wasserstein gradient flow of the global relative entropy from Fokker-Planck equation. Compared with existing schemes, Wasserstein gradient flow is a smoother and near-optimal numerical scheme to approximate real data densities.
  arXiv  Detail & Related papers  (2019-10-31T02:26:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.