Open Access. Powered by Scholars. Published by Universities.®
Physical Sciences and Mathematics Commons™
Open Access. Powered by Scholars. Published by Universities.®
- Discipline
-
- Computer Sciences (9)
- Engineering (3)
- Artificial Intelligence and Robotics (2)
- Aerospace Engineering (1)
- Chemistry (1)
-
- Civil and Environmental Engineering (1)
- Clinical Psychology (1)
- Electrical and Computer Engineering (1)
- Materials Science and Engineering (1)
- Mechanical Engineering (1)
- Medicinal-Pharmaceutical Chemistry (1)
- Multi-Vehicle Systems and Air Traffic Control (1)
- Operations Research, Systems Engineering and Industrial Engineering (1)
- Physics (1)
- Psychology (1)
- Social and Behavioral Sciences (1)
- Software Engineering (1)
- Keyword
-
- Machine Learning (2)
- 3D Classification (1)
- Asset Management (1)
- Azurin (1)
- Continual Learning (1)
-
- DNA (1)
- Data visualization (1)
- Deep Learning (1)
- Dementia (1)
- Embedded Devices (1)
- Evolutionary algorithm (1)
- Extended reality (1)
- Federated Learning (1)
- Few-Shot Learning (1)
- Foreign language learning (1)
- G-Quadruplexes (1)
- Genetic algorithm (1)
- Hazardous Waste (1)
- Human-computer interaction (1)
- Immersive analytics (1)
- Infrastructure (1)
- Interactivity (1)
- Ionic Liquids (1)
- Machine learning (1)
- Meta Learning (1)
- Mild Cognitive Impairment (1)
- Molecular Dynamics Simulations (1)
- Multi-view (1)
- Neural architecture engineering (1)
- Neural network (1)
Articles 1 - 11 of 11
Full-Text Articles in Physical Sciences and Mathematics
Investigating The Effects Of Ionic Liquids On Dna Gquadruplex And Protein Structure Using Molecular Dynamics Simulations, Nicholas J. Paradis
Investigating The Effects Of Ionic Liquids On Dna Gquadruplex And Protein Structure Using Molecular Dynamics Simulations, Nicholas J. Paradis
Theses and Dissertations
Nucleic acids and proteins have huge implications in biomedicine and bioengineering, however their storage instability limits their applicability and current storage protocols are expensive and globally-inaccessible. Finding an alternative biocompatible media to store nucleic acids and proteins would reduce costs and increase their applicability. Ionic liquids (ILs) are molten salt compounds that have been shown to modulate the stability and activity of nucleic acids and proteins. In this thesis, molecular modeling studies of DNA/RNA and protein structure in ILs will be discussed (Chapter 1) and this method will be used to study the IL effects on the structure on the …
Utilizing Federated Learning And Meta Learning For Few-Shot Learning On Edge Devices, Kousalya Soumya Lahari Voleti
Utilizing Federated Learning And Meta Learning For Few-Shot Learning On Edge Devices, Kousalya Soumya Lahari Voleti
Theses and Dissertations
The efficient and effective handling of few-shot learning tasks on mobile devices is challenging due to the small training set issue and the physical limitations in power and computational resources on these devices. In this thesis, we propose a solution that combines federated learning and meta-learning to handle independent few-shot learning tasks on multiple devices (or clients) and the server. In particular, we utilize the Prototypical Networks to perform meta-learning on all devices to learn multiple independent few-shot learning models and to combine the models in a centralized data distributed architecture using federated learning which can be reused by the …
Investigation Of Adhesion, Deformation Mechanics, And Electrical Properties Of Ag/Sio2/Pdms Tri-Layers For Stretchable Electronic Applications, Rhandy Joe Paladines
Investigation Of Adhesion, Deformation Mechanics, And Electrical Properties Of Ag/Sio2/Pdms Tri-Layers For Stretchable Electronic Applications, Rhandy Joe Paladines
Theses and Dissertations
The motivation behind this research is to improve the interfacial layer bonding of metallic thin films to PDMS substrates with the aid of a buffer layer. The physical vapor deposition (PVD) technique of sputtering was used to deposit bilayer thin films of silver (Ag) and silicon dioxide (SiO2) on PDMS. Two chamber pressures were used in this work, 5 and 20 mTorr, to investigate the role of this parameter in determining the interfacial adhesion and the role in determining the resistance sensitivity. Studies of the surface energy and increased bonding strength for metallization are carried out. Surface characterization using atomic …
A Machine Learning Framework For Automatic Speech Recognition In Air Traffic Control Using Word Level Binary Classification And Transcription, Fowad Shahid Sohail
A Machine Learning Framework For Automatic Speech Recognition In Air Traffic Control Using Word Level Binary Classification And Transcription, Fowad Shahid Sohail
Theses and Dissertations
Advances in Artificial Intelligence and Machine learning have enabled a variety of new technologies. One such technology is Automatic Speech Recognition (ASR), where a machine is given audio and transcribes the words that were spoken. ASR can be applied in a variety of domains to improve general usability and safety. One such domain is Air Traffic Control (ATC). ASR in ATC promises to improve safety in a mission critical environment. ASR models have historically required a large amount of clean training data. ATC environments are noisy and acquiring labeled data is a difficult, expertise dependent task. This thesis attempts to …
Effective Immersive Analytics For Everyday Use, Benjamin D. Weidner
Effective Immersive Analytics For Everyday Use, Benjamin D. Weidner
Theses and Dissertations
Data visualization is an important field of work that takes in uncountable amounts of indexes to create an easy-to-read interpretation of what was previously unreadable. Immersive analytics is the new field that brings 3D data visualization to virtual reality, immersing users directly into the data. Focusing on bringing humans and computers closer together through natural function can benefit the world of data science. In order to accurately utilize this field to benefit this world, principles must be laid out and observed to see which techniques and methods are best fit for an everyday immersive analytics platform. Our findings show that, …
Development Of Software Tools For Efficient And Sustainable Process Development And Improvement, Jake P. Stengel
Development Of Software Tools For Efficient And Sustainable Process Development And Improvement, Jake P. Stengel
Theses and Dissertations
Infrastructure is a key component in the well-being of our society that leads to its growth, development, and productive operations. A well-built infrastructure allows the community to be more competitive and promotes economic advancement. In 2021, the ASCE (American Society of Civil Engineers) ranked the American infrastructure as substandard, with an overall grade of C-. The overall ranking suffers when key infrastructure categories are not maintained according to the needs of the population. Therefore, there is a need to consider alternative methods to improve our infrastructure and make it more sustainable to enhance the overall grade. One of the challenges …
An Empirical Study On Sampling Approaches For 3d Image Classification Using Deep Learning, Nicholas Michelette
An Empirical Study On Sampling Approaches For 3d Image Classification Using Deep Learning, Nicholas Michelette
Theses and Dissertations
A 3D classification method requires more training data than a 2D image classification method to achieve good performance. These training data usually come in the form of multiple 2D images (e.g., slices in a CT scan) or point clouds (e.g., 3D CAD modeling) for volumetric object representation. The amount of data required to complete this higher dimension problem comes with the cost of requiring more processing time and space. This problem can be mitigated with data size reduction (i.e., sampling). In this thesis, we empirically study and compare the classification performance and deep learning training time of PointNet utilizing uniform …
A Digital Application For Assessment Of Neurocognitive Disabilities, Thomas H. Auriemma
A Digital Application For Assessment Of Neurocognitive Disabilities, Thomas H. Auriemma
Theses and Dissertations
Background: Neuropsychological assessment is designed to identify neurocognitive impairment and has traditionally relied on pen-and-paper tests. The behavior collected from these tests is usually expressed as a total summary score or a score that reflects a restricted number of features that assess errors. There is now interest in coupling traditional paper and pencil tests with digital assessment technology. In this context traditional metrics such as summary scores are still available. However, using digital technology, a host of time-based parameters can now be obtained. These time-based parameters include the total time to complete the task or total time to completion, as …
Assessing The Effect Of Interactivity On Virtual Reality Second Language Learning, Christene Harris
Assessing The Effect Of Interactivity On Virtual Reality Second Language Learning, Christene Harris
Theses and Dissertations
Virtual Reality (VR) being used as a helpful tool in language education is widely supported by the current literature. It can provide a variety of stimulating scenarios that keep learner engagement high. The use of VR for language learning is a research area that has shown promise in recent years. This makes it necessary for further research to be conducted in the field to determine ways to maximize its potential. This thesis aims to determine if the level of interactivity present in a VR Language Learning Application is a factor that will impact a user's capability to successfully learn a …
Low Memory Continual Learning Classification Algorithms For Low Resource Hardware, Autumn Lilly Chadwick
Low Memory Continual Learning Classification Algorithms For Low Resource Hardware, Autumn Lilly Chadwick
Theses and Dissertations
Continual Learning (CL) is a machine learning approach which focuses on continuous learning of data rather than single dataset-based learning. In this thesis, this same focus is applied with respect to the field of machine learning for embedded devices which is still in the early stages of development. This focus is further used to develop various algorithms such as utilizing prior trained starting networks, weighted output schemes, and replay or reduced datasets for training while maintaining a consistent focus on low resource devices to maintain acceptable performance. The experimental results show an improvement in model training times as compared to …
An Empirical Study On The Efficacy Of Evolutionary Algorithms For Automated Neural Architecture Search, Andrew D. Cuccinello
An Empirical Study On The Efficacy Of Evolutionary Algorithms For Automated Neural Architecture Search, Andrew D. Cuccinello
Theses and Dissertations
The configuration and architecture design of neural networks is a time consuming process that has been shown to provide significant training speed and prediction improvements. Traditionally, this process is done manually, but this requires a large amount of expert knowledge and significant investment of labor. As a result it is beneficial to have automated ways to optimize model architectures. In this thesis, we study the use of evolutionary algorithm for neural architecture search (NAS). Moreover, we investigate the effect of integrating evolutionary NAS into deep reinforcement learning to learn control policy for ATARI game playing. Empirical classification results on the …