Open Access. Powered by Scholars. Published by Universities.®
Physical Sciences and Mathematics Commons™
Open Access. Powered by Scholars. Published by Universities.®
Articles 1 - 9 of 9
Full-Text Articles in Physical Sciences and Mathematics
Leveraging Subject Matter Expertise To Optimize Machine Learning Techniques For Air And Space Applications, Philip Y. Cho
Leveraging Subject Matter Expertise To Optimize Machine Learning Techniques For Air And Space Applications, Philip Y. Cho
Theses and Dissertations
We develop new machine learning and statistical methods that are tailored for Air and Space applications through the incorporation of subject matter expertise. In particular, we focus on three separate research thrusts that each represents a different type of subject matter knowledge, modeling approach, and application. In our first thrust, we incorporate knowledge of natural phenomena to design a neural network algorithm for localizing point defects in transmission electron microscopy (TEM) images of crystalline materials. In our second research thrust, we use Bayesian feature selection and regression to analyze the relationship between fighter pilot attributes and flight mishap rates. We …
Application Of Machine Learning Models With Numerical Simulations Of An Experimental Microwave Induced Plasma Gasification Reactor, Owen D. Sedej
Application Of Machine Learning Models With Numerical Simulations Of An Experimental Microwave Induced Plasma Gasification Reactor, Owen D. Sedej
Theses and Dissertations
This thesis aims to contribute to the future development of this technology by providing an in-depth literature review of how this technology physically operates and can be numerically modeled. Additionally, this thesis reviews literature of machine learning models that have been applied to gasification to make accurate predictions regarding the system. Finally, this thesis provides a framework of how to numerically model an experimental plasma gasification reactor in order to inform a variety of machine learning models.
Analysis Of Generalized Artificial Intelligence Potential Through Reinforcement And Deep Reinforcement Learning Approaches, Jonathan Turner
Analysis Of Generalized Artificial Intelligence Potential Through Reinforcement And Deep Reinforcement Learning Approaches, Jonathan Turner
Theses and Dissertations
Artificial Intelligence is the next competitive domain; the first nation to develop human level artificial intelligence will have an impact similar to the development of the atomic bomb. To maintain the security of the United States and her people, the Department of Defense has funded research into the development of artificial intelligence and its applications. This research uses reinforcement learning and deep reinforcement learning methods as proxies for current and future artificial intelligence agents and to assess potential issues in development. Agent performance were compared across two games and one excursion: Cargo Loading, Tower of Hanoi, and Knapsack Problem, respectively. …
Automated Aircraft Visual Inspection With Artificial Data Generation Enabled Deep Learning, Nathan J. Gaul
Automated Aircraft Visual Inspection With Artificial Data Generation Enabled Deep Learning, Nathan J. Gaul
Theses and Dissertations
Aircraft visual inspection, which is essential to daily maintenance of an aircraft, is expensive and time-consuming to perform. Augmenting trained maintenance technicians with automated UAVs to collect and analyze images for aircraft inspection is an active research topic and a potential application of CNNs. Training datasets for niche research topics such as aircraft visual inspection are small and challenging to produce, and the manual process of labeling these datasets often produces subjective annotations. Recently, researchers have produced several successful applications of artificially generated datasets with domain randomization for training CNNs for real-world computer vision problems. The research outlined herein builds …
Characterizing Convolutional Neural Network Early-Learning And Accelerating Non-Adaptive, First-Order Methods With Localized Lagrangian Restricted Memory Level Bundling, Benjamin O. Morris
Characterizing Convolutional Neural Network Early-Learning And Accelerating Non-Adaptive, First-Order Methods With Localized Lagrangian Restricted Memory Level Bundling, Benjamin O. Morris
Theses and Dissertations
This dissertation studies the underlying optimization problem encountered during the early-learning stages of convolutional neural networks and introduces a training algorithm competitive with existing state-of-the-art methods. First, a Design of Experiments method is introduced to systematically measure empirical second-order Lipschitz upper bound and region size estimates for local regions of convolutional neural network loss surfaces experienced during the early-learning stages. This method demonstrates that architecture choices can significantly impact the local loss surfaces traversed during training. Next, a Design of Experiments method is used to study the effects convolutional neural network architecture hyperparameters have on different optimization routines' abilities to …
Algorithm Selection Framework: A Holistic Approach To The Algorithm Selection Problem, Marc W. Chalé
Algorithm Selection Framework: A Holistic Approach To The Algorithm Selection Problem, Marc W. Chalé
Theses and Dissertations
A holistic approach to the algorithm selection problem is presented. The “algorithm selection framework" uses a combination of user input and meta-data to streamline the algorithm selection for any data analysis task. The framework removes the conjecture of the common trial and error strategy and generates a preference ranked list of recommended analysis techniques. The framework is performed on nine analysis problems. Each of the recommended analysis techniques are implemented on the corresponding data sets. Algorithm performance is assessed using the primary metric of recall and the secondary metric of run time. In six of the problems, the recall of …
Parallelization Of Ant Colony Optimization Via Area Of Expertise Learning, Adrian A. De Freitas
Parallelization Of Ant Colony Optimization Via Area Of Expertise Learning, Adrian A. De Freitas
Theses and Dissertations
Ant colony optimization algorithms have long been touted as providing an effective and efficient means of generating high quality solutions to NP-hard optimization problems. Unfortunately, while the structure of the algorithm is easy to parallelize, the nature and amount of communication required for parallel execution has meant that parallel implementations developed suffer from decreased solution quality, slower runtime performance, or both. This thesis explores a new strategy for ant colony parallelization that involves Area of Expertise (AOE) learning. The AOE concept is based on the idea that individual agents tend to gain knowledge of different areas of the search space …
Discovery Learning In Autonomous Agents Using Genetic Algorithms, Edward O. Gordon
Discovery Learning In Autonomous Agents Using Genetic Algorithms, Edward O. Gordon
Theses and Dissertations
As the new Distributed Interactive Simulation (DIS) draft standard evolves into a useful document and distributed simulations begin to emerge that implement parts of the standard, there is renewed interest in available methods to effectively control autonomous aircraft agents in such a simulated environment. This investigation examines the use of a genetics-based classifier system for agent control. These are robust learning systems that use the adaptive search mechanisms of genetic algorithms to guide the learning system in forming new concepts (decision rules) about its environment. By allowing the rule base to evolve, it adapts agent behavior to environmental changes. Addressed …
Using Discovery-Based Learning To Prove The Behavior Of An Autonomous Agent, David N. Mezera
Using Discovery-Based Learning To Prove The Behavior Of An Autonomous Agent, David N. Mezera
Theses and Dissertations
Computer-generated autonomous agents in simulation often behave predictably and unrealistically. These characteristics make them easy to spot and exploit by human participants in the simulation, when we would prefer the behavior of the agent to be indistinguishable from human behavior. An improvement in behavior might be possible by enlarging the library of responses, giving the agent a richer assortment of tactics to employ during a combat scenario. Machine learning offers an exciting alternative to constructing additional responses by hand by instead allowing the system to improve its own performance with experience. This thesis presents NOSTRUM, a discovery-based learning DBL system …