Options
From feature selection to neural architecture search : development and implementation of AI-based algorithms to enhance AI-driven research
Citation Link: https://doi.org/10.15480/882.13706
Publikationstyp
Doctoral Thesis
Date Issued
2024
Sprache
English
Author(s)
Advisor
Referee
Title Granting Institution
Technische Universität Hamburg
Place of Title Granting Institution
Hamburg
Examination Date
2024-09-19
Institute
TORE-DOI
Citation
Technische Universität Hamburg (2024)
Effectively applying machine learning methods, particularly in applied sciences, can pose significant challenges.
However, when employed correctly, these algorithms prove to be powerful tools, offering substantial benefits across a wide range of research applications.
Fine-tuning them to individual needs and circumstances requires making a number of relevant and well-informed choices, all of which can profoundly impact the quality of the outcome.
In this thesis, I present a comprehensive overview over the machine learning process, along with two use-cases of successful machine learning application in practice.
However, when employed correctly, these algorithms prove to be powerful tools, offering substantial benefits across a wide range of research applications.
Fine-tuning them to individual needs and circumstances requires making a number of relevant and well-informed choices, all of which can profoundly impact the quality of the outcome.
In this thesis, I present a comprehensive overview over the machine learning process, along with two use-cases of successful machine learning application in practice.
Subjects
machine learning
deep learning
feature selection
neural architecture search
optimization
automation
DDC Class
006.3: Artificial Intelligence
Loading...
Name
Schiessler_EJ-from_feature_selection_to_neural_architecture_search.pdf
Size
12.63 MB
Format
Adobe PDF