Search Results for author: Daniel Livescu

Found 10 papers, 2 papers with code

Compressed Convolutional LSTM: An Efficient Deep Learning framework to Model High Fidelity 3D Turbulence

no code implementations28 Feb 2019 Arvind Mohan, Don Daniel, Michael Chertkov, Daniel Livescu

High-fidelity modeling of turbulent flows is one of the major challenges in computational physics, with diverse applications in engineering, earth sciences and astrophysics, among many others.

Fluid Dynamics Chaotic Dynamics Computational Physics

Leveraging Bayesian Analysis To Improve Accuracy of Approximate Models

no code implementations20 May 2019 Balasubramanya T. Nadiga, Chiyu Jiang, Daniel Livescu

We focus on improving the accuracy of an approximate model of a multiscale dynamical system that uses a set of parameter-dependent terms to account for the effects of unresolved or neglected dynamics on resolved scales.

Uncertainty Quantification Variational Inference

Embedding Hard Physical Constraints in Neural Network Coarse-Graining of 3D Turbulence

no code implementations31 Jan 2020 Arvind T. Mohan, Nicholas Lubbers, Daniel Livescu, Michael Chertkov

In the recent years, deep learning approaches have shown much promise in modeling complex systems in the physical sciences.

Computational Physics

Wavelet-Powered Neural Networks for Turbulence

no code implementations ICLR Workshop DeepDiffEq 2019 Arvind T. Mohan, Daniel Livescu, Michael Chertkov

One of the fundamental driving phenomena for applications in engineering, earth sciences and climate is fluid turbulence.

Objective discovery of dominant dynamical processes with intelligible machine learning

1 code implementation21 Jun 2021 Bryan E. Kaiser, Juan A. Saenz, Maike Sonnewald, Daniel Livescu

The advent of big data has vast potential for discovery in natural phenomena ranging from climate science to medicine, but overwhelming complexity stymies insight.

BIG-bench Machine Learning Dimensionality Reduction

Regression-based projection for learning Mori-Zwanzig operators

no code implementations10 May 2022 Yen Ting Lin, Yifeng Tian, Danny Perez, Daniel Livescu

We propose to adopt statistical regression as the projection operator to enable data-driven learning of the operators in the Mori--Zwanzig formalism.

regression

Predictive Scale-Bridging Simulations through Active Learning

no code implementations20 Sep 2022 Satish Karra, Mohamed Mehana, Nicholas Lubbers, Yu Chen, Abdourahmane Diaw, Javier E. Santos, Aleksandra Pachalieva, Robert S. Pavel, Jeffrey R. Haack, Michael McKerns, Christoph Junghans, Qinjun Kang, Daniel Livescu, Timothy C. Germann, Hari S. Viswanathan

Throughout computational science, there is a growing need to utilize the continual improvements in raw computational horsepower to achieve greater physical fidelity through scale-bridging over brute-force increases in the number of mesh elements.

Active Learning

Physics-Constrained Generative Adversarial Networks for 3D Turbulence

no code implementations1 Dec 2022 Dima Tretiak, Arvind T. Mohan, Daniel Livescu

Generative Adversarial Networks (GANs) have received wide acclaim among the machine learning (ML) community for their ability to generate realistic 2D images.

Cannot find the paper you are looking for? You can Submit a new open access paper.