Skip to Main Content
Frequently Asked Questions
Submit an ETD
Global Search Box
Need Help?
Keyword Search
Participating Institutions
Advanced Search
School Logo
Files
File List
Dissertation_Wei_Hung_Su_Final.pdf (2.5 MB)
ETD Abstract Container
Abstract Header
Utilizing Data-Driven Modeling to Characterize Biological Systems
Author Info
Su, Wei-Hung
Permalink:
http://rave.ohiolink.edu/etdc/view?acc_num=osu1681941476945444
Abstract Details
Year and Degree
2023, Doctor of Philosophy, Ohio State University, Mathematics.
Abstract
This dissertation presents three novel methodologies for Data-Driven modeling to biological systems with complex dynamical behavior. First of all, a numerical procedure based on deep learning methods, particularly residual, and flow-map learning tools for dynamical systems is presented to accurately construct numerical dynamic system from measurement data and incorporate unknown parameters. The trained deep neural network model can be used as a predictive tool to produce system predictions of different settings and conduct detailed analyses of the underlying process. In this section, three different biological models are utilized to display the performance of the proposed method. Secondly, a deep neural network structure with memory terms and recurrent structures is proposed for modeling biological systems with partially observed data containing hidden parameters. Through several representative biological problems, the methodology is shown to capture qualitative dynamical behavior changes in the system, even when the parameters controlling such changes are completely unknown. The learned deep neural network model effectively creates a "closed" model involving only the observables when such a closed-form model does not exist mathematically. Last but not least, the focus shifts towards particle movement modeling, and the proposed coarse density methodology is used to identify the governing equation of the density dynamic of particle systems. By utilizing the recurrent DNN with memory terms, the methodology is shown to be effective in recovering the underlying governing equations in the second part of the thesis. The flocking and traffic models are used to showcase the methodology and demonstrate how the learned deep neural network can represent the density dynamic of the original model. Furthermore, the relation between the Lagrange and Euler framework is also discussed using the traffic model.
Committee
Dongbin Xiu (Advisor)
Janet Best (Committee Member)
Yulong Xing (Committee Member)
Subject Headings
Mathematics
Recommended Citations
Refworks
EndNote
RIS
Mendeley
Citations
Su, W.-H. (2023).
Utilizing Data-Driven Modeling to Characterize Biological Systems
[Doctoral dissertation, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu1681941476945444
APA Style (7th edition)
Su, Wei-Hung.
Utilizing Data-Driven Modeling to Characterize Biological Systems.
2023. Ohio State University, Doctoral dissertation.
OhioLINK Electronic Theses and Dissertations Center
, http://rave.ohiolink.edu/etdc/view?acc_num=osu1681941476945444.
MLA Style (8th edition)
Su, Wei-Hung. "Utilizing Data-Driven Modeling to Characterize Biological Systems." Doctoral dissertation, Ohio State University, 2023. http://rave.ohiolink.edu/etdc/view?acc_num=osu1681941476945444
Chicago Manual of Style (17th edition)
Abstract Footer
Document number:
osu1681941476945444
Download Count:
0
Copyright Info
© 2023, some rights reserved.
Utilizing Data-Driven Modeling to Characterize Biological Systems by Wei-Hung Su is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License. Based on a work at etd.ohiolink.edu.
This open access ETD is published by The Ohio State University and OhioLINK.