Date of Award
3-2024
Document Type
Thesis
Degree Name
Master of Science in Operations Research
Department
Department of Operational Sciences
First Advisor
Lance E. Champagne, PhD
Abstract
Deep neural networks and transfer learning show potential in addressing complex problems such as the Tower of Hanoi and knapsack problems. The primary aim is to examine how the use of deep neural networks and transfer learning can enhance the ability of artificial learning systems to generalize. Transfer learning plays a crucial role in machine learning, particularly in the domain of artificial neural networks, as it helps overcome the challenges associated with limited data, computational efficiency, and generalization. The methodology used in this research involves the creation of data sets for the Tower of Hanoi and knapsack problems. To predict solution sequences and optimize problem solving strategies, a combination of convolutional neural networks (CNNs) and long-short-term memory (LSTM) models is used. The findings demonstrate that the CNN model effectively identifies recursive patterns in the Tower of Hanoi problem. Furthermore, a hybrid CNN-LSTM model shows promise in solving the knapsack problem, although it encounters difficulties in learning capacity constraints. Consequently, a post-processing step is necessary to ensure compliance with these constraints.
AFIT Designator
AFIT-ENS-MS-24-M-086
Recommended Citation
Lang, Jacob S., "The Use of Deep Learning and Transfer Learning in Complex Problems" (2024). Theses and Dissertations. 7721.
https://scholar.afit.edu/etd/7721
Comments
A 12-month embargo was observed for posting this work on AFIT Scholar.
Distribution Statement A, Approved for Public Release. PA case number on file.