Back to search

Institutes outside Greece

Physics-informed numerical analysis and machine learning for modelling and analysis of large scale complex systems with applications in computational neuroscience

Abstract

dc:description

Τhis Ph.D. Thesis deals with the development and implementation of physics-informed numerical analysis and machine learning methods for the solution of both the inverse and forward problem in complex systems modeling including the numerical bifurcation analysis of the emergent dynamics from big data. Modeling and analyzing the emergent behavior of large-scale complex dynamical systems from experimental data and/or data produced by detailed high-fidelity microscopic simulations requires appropriate data-driven/numerical analysis-based methods for extracting coarse-scale models that can be utilized for further numerical analysis of the emergent dynamics. The springs of this Thesis are interdisciplinary, bridging state-of-the-art methodologies from numerical analysis, machine learning, microscopic simulations, bifurcation theory, and neuroimaging. The research efforts and results were focused in three main directions. We have first focused on the solution of the inverse problem of source-localization in neuroimaging, exploiting and comparing the performance of state-of-the-art regularization methods, namely the standardized Low Resolution Electromagnetic Tomography (sLORETA), the weighted Minimum Norm Estimation (wMNE) and the dynamic Statistical Parametric Mapping (dSPM) and information arising from the electrophysiology of the brain. In particular, the research efforts were focused on the localization of the sources of brain activity of children with epilepsy based on electroencephalography (EEG) recordings acquired during a visual discrimination working memory (WM) task using numerical regularization algorithms. Importantly, our study and findings reveal also the importance and potential that originates from the use of physics-based information as well as publicly available scientific resources such as the “Neurodevelopmental MRI" database, which allows the researchers to numerically analyze available neuroimaging data and investigate questions beyond the scope of the original studies. Next, we addressed a computational framework for the solution of the inverse problem of complex systems modeling, i.e. the construction of PDEs from spatiotemporal data. Towards this aim, we (a) first embed the high-dimensional spatiotemporal data produced by microscopic simulators in a low-dimensional manifold with the aid of manifold learning algorithms and in particular Diffusion Maps, by identifying an appropriate set of parsimonious observables, (b) based on the extracted coarse-observables, we use machine learning to identify an effective PDE and, (c) based on the PDE model, we solve the forward problem, thus we solve and construct the corresponding coarse-grained bifurcation diagram in order to systematically study the emergent dynamics. For our illustrations, we implemented the proposed method to construct the one-dimensional bifurcation diagram of the celebrated FitzHugh-Nagumo PDEs of activation-inhibition neuronal dynamics from data generated by Lattice Boltzmann simulations. Finally, we addressed a numerical method based on physics-informed random-projection neural networks to solve the forward problem in initial value problems (IVPs) of systems of ODEs (which may also arise from the spatial discretization ofPDEs). The numerical solution of the IVPs is obtained by constructing a system of nonlinear algebraic equations, which is solved with respect to the output weights by Newton’s iterations. The performance of the proposed scheme was assessed through three benchmark stiff IVPs, namely the Prothero-Robinson, the van der Pol model, and the ROBER problem. Furthermore, the proposed scheme was compared with an adaptive Runge-Kutta method, and a variable-step variable-order multistep solver based on numerical differentiation formulas, as implemented in the ode45 and ode15s of the Matlab ODE suite. We show that the proposed scheme yields good approximation accuracy, thus outperforming in some cases ode45 and ode15s, especially in the cases where steep gradients arise. Furthermore, the computational times of our approach are comparable with those of the two Matlab solvers for all practical purposes.

Degree

thesis:*
Grantor dc:publisher
Institutes outside Greece
Year dc:date
2022

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Galaris, Evangelos
  • Γαλάρης, Ευάγγελος

Subjects

dc:subject × 16

Rights

Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/55817

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Galaris, Evangelos; Γαλάρης, Ευάγγελος. Physics-informed numerical analysis and machine learning for modelling and analysis of large scale complex systems with applications in computational neuroscience. Institutes outside Greece, 2022. http://hdl.handle.net/10442/hedi/55817