Loading...
Please wait, while we are loading the content...
Similar Documents
Gaussian Process Regression for Maximum Entropy Distribution
| Content Provider | arXiv |
|---|---|
| Author | Sadr, Mohsen Torrilhon, Manuel Gorji, M. Hossein |
| Date of Submission | 2023-08-11 |
| Abstract | Maximum-Entropy Distributions offer an attractive family of probability densities suitable for moment closure problems. Yet finding the Lagrange multipliers which parametrize these distributions, turns out to be a computational bottleneck for practical closure settings. Motivated by recent success of Gaussian processes, we investigate the suitability of Gaussian priors to approximate the Lagrange multipliers as a map of a given set of moments. Examining various kernel functions, the hyperparameters are optimized by maximizing the log-likelihood. The performance of the devised data-driven Maximum-Entropy closure is studied for couple of test cases including relaxation of non-equilibrium distributions governed by Bhatnagar-Gross-Krook and Boltzmann kinetic equations. |
| Related Links | https://arxiv.org/pdf/2308.06149.pdf |
| arXiv | 2308.06149 |
| Language | English |
| Access Restriction | Open |
| Subject Keyword | Statistics - Machine Learning Computer Science - Machine Learning Mathematical Physics Physics - Data Analysis, Statistics and Probability Statistics Physics Computer Science Mathematics |
| Content Type | Text |
| Resource Type | Article |
| Subject | Statistics and Probability |
