Zobrazeno 1 - 10
of 40
pro vyhledávání: '"Kontolati, Katiana"'
Autor:
Kumar, Varun, Goswami, Somdatta, Kontolati, Katiana, Shields, Michael D., Karniadakis, George Em
Multi-task learning (MTL) is an inductive transfer mechanism designed to leverage useful information from multiple tasks to improve generalization performance compared to single-task learning. It has been extensively explored in traditional machine l
Externí odkaz:
http://arxiv.org/abs/2408.02198
Autor:
Tsapetis, Dimitrios, Shields, Michael D., Giovanis, Dimitris G., Olivier, Audrey, Novak, Lukas, Chakroborty, Promit, Sharma, Himanshu, Chauhan, Mohit, Kontolati, Katiana, Vandanapu, Lohit, Loukrezis, Dimitrios, Gardner, Michael
This paper presents the latest improvements introduced in Version 4 of the UQpy, Uncertainty Quantification with Python, library. In the latest version, the code was restructured to conform with the latest Python coding conventions, refactored to sim
Externí odkaz:
http://arxiv.org/abs/2305.09572
Operator regression provides a powerful means of constructing discretization-invariant emulators for partial-differential equations (PDEs) describing physical systems. Neural operators specifically employ deep neural networks to approximate mappings
Externí odkaz:
http://arxiv.org/abs/2304.07599
Transfer learning (TL) enables the transfer of knowledge gained in learning to perform one task (source) to a related but different task (target), hence addressing the expense of data acquisition and labeling, potential computational power limitation
Externí odkaz:
http://arxiv.org/abs/2204.09810
Constructing accurate and generalizable approximators for complex physico-chemical processes exhibiting highly non-smooth dynamics is challenging. In this work, we propose new developments and perform comparisons for two promising approaches: manifol
Externí odkaz:
http://arxiv.org/abs/2203.05071
Autor:
Kontolati, Katiana, Loukrezis, Dimitrios, Giovanis, Dimitris G., Vandanapu, Lohit, Shields, Michael D.
Constructing surrogate models for uncertainty quantification (UQ) on complex partial differential equations (PDEs) having inherently high-dimensional $\mathcal{O}(10^{\ge 2})$ stochastic inputs (e.g., forcing terms, boundary conditions, initial condi
Externí odkaz:
http://arxiv.org/abs/2202.04648
Autor:
Santos, Ketson R. M. dos, Giovanis, Dimitrios G., Kontolati, Katiana, Loukrezis, Dimitrios, Shields, Michael D.
In this paper, a novel surrogate model based on the Grassmannian diffusion maps (GDMaps) and utilizing geometric harmonics is developed for predicting the response of engineering systems and complex physical phenomena. The method utilizes the GDMaps
Externí odkaz:
http://arxiv.org/abs/2109.13805
Constructing probability densities for inference in high-dimensional spectral data is often intractable. In this work, we use normalizing flows on structured spectral latent spaces to estimate such densities, enabling downstream inference tasks. In a
Externí odkaz:
http://arxiv.org/abs/2108.08709
Autor:
Kontolati, Katiana, Loukrezis, Dimitrios, Santos, Ketson R. M. dos, Giovanis, Dimitrios G., Shields, Michael D.
In this work we introduce a manifold learning-based method for uncertainty quantification (UQ) in systems describing complex spatiotemporal processes. Our first objective is to identify the embedding of a set of high-dimensional data representing qua
Externí odkaz:
http://arxiv.org/abs/2107.09814
Autor:
Kontolati, Katiana, Alix-Williams, Darius, Boffi, Nicholas M., Falk, Michael L., Rycroft, Chris H., Shields, Michael D.
We introduce a generalized machine learning framework to probabilistically parameterize upper-scale models in the form of nonlinear PDEs consistent with a continuum theory, based on coarse-grained atomistic simulation data of mechanical deformation a
Externí odkaz:
http://arxiv.org/abs/2103.00779