Open Access. Powered by Scholars. Published by Universities.®
Physical Sciences and Mathematics Commons™
Open Access. Powered by Scholars. Published by Universities.®
Articles 1 - 17 of 17
Full-Text Articles in Physical Sciences and Mathematics
Open System Neural Networks, Bradley Hatch
Open System Neural Networks, Bradley Hatch
Theses and Dissertations
Recent advances in self-supervised learning have made it possible to reuse information-rich models that have been generally pre-trained on massive amounts of data for other downstream tasks. But the pre-training process can be drastically different from the fine-tuning training process, which can lead to inefficient learning. We address this disconnect in training dynamics by structuring the learning process like an open system in thermodynamics. Open systems can achieve a steady state when low-entropy inputs are converted to high-entropy outputs. We modify the the model and the learning process to mimic this behavior, and attend more to elements of the input …
More Is Better Than One: The Effect Of Ensembling On Deep Learning Performance In Biochemical Prediction Problems, Jacob A. Stern
More Is Better Than One: The Effect Of Ensembling On Deep Learning Performance In Biochemical Prediction Problems, Jacob A. Stern
Theses and Dissertations
This thesis presents two papers addressing important biochemical prediction challenges. The first paper focuses on accurate protein distance predictions and introduces updates to the ProSPr network. We evaluate its performance in the Critical Assessment of techniques for Protein Structure Prediction (CASP14) competition, investigating its accuracy dependence on sequence length and multiple sequence alignment depth. The ProSPr network, an ensemble of three convolutional neural networks (CNNs), demonstrates superior performance compared to individual networks. The second paper addresses the issue of accurate ligand ranking in virtual screening for drug discovery. We propose MILCDock, a machine learning consensus docking tool that leverages predictions …
Zero And Few-Shot Concept Learning With Pre-Trained Embeddings, Jamison M. Moody
Zero And Few-Shot Concept Learning With Pre-Trained Embeddings, Jamison M. Moody
Theses and Dissertations
Neural networks typically struggle with reasoning tasks on out of domain data, something that humans can more easily adapt to. Humans come with prior knowledge of concepts and can segment their environment into building blocks (such as objects) that allow them to reason effectively in unfamiliar situations. Using this intuition, we train a network that utilizes fixed embeddings from the CLIP (Contrastive Language--Image Pre-training) model to do a simple task that the original CLIP model struggles with. The network learns concepts (such as "collide" and "avoid") in a supervised source domain in such a way that the network can adapt …
Language Modeling Using Image Representations Of Natural Language, Seong Eun Cho
Language Modeling Using Image Representations Of Natural Language, Seong Eun Cho
Theses and Dissertations
This thesis presents training of an end-to-end autoencoder model using the transformer, with an encoder that can encode sentences into fixed-length latent vectors and a decoder that can reconstruct the sentences using image representations. Encoding and decoding sentences to and from these image representations are central to the model design. This method allows new sentences to be generated by traversing the Euclidean space, which makes vector arithmetic possible using sentences. Machines excel in dealing with concrete numbers and calculations, but do not possess an innate infrastructure designed to help them understand abstract concepts like natural language. In order for a …
Towards Color-Based Two-Hand 3d Global Pose Estimation, Fanqing Lin
Towards Color-Based Two-Hand 3d Global Pose Estimation, Fanqing Lin
Theses and Dissertations
Pose estimation and tracking is essential for applications involving human controls. Specifically, as the primary operating tool for human activities, hand pose estimation plays a significant role in applications such as hand tracking, gesture recognition, human-computer interaction and VR/AR. As the field develops, there has been a trend to utilize deep learning to estimate the 2D/3D hand poses using color-based information without depth data. Within the depth-based as well as color-based approaches, the research community has primarily focused on single-hand scenarios in a localized/normalized coordinate system. Due to the fact that both hands are utilized in most applications, we propose …
Language Learning Using Models Of Intentionality In Repeated Games With Cheap Talk, Jonathan Berry Skaggs
Language Learning Using Models Of Intentionality In Repeated Games With Cheap Talk, Jonathan Berry Skaggs
Theses and Dissertations
Language is critical to establishing long-term cooperative relationships among intelligent agents (including people), particularly when the agents' preferences are in conflict. In such scenarios, an agent uses speech to coordinate and negotiate behavior with its partner(s). While recent work has shown that neural language modeling can produce effective speech agents, such algorithms typically only accept previous text as input. However, in relationships among intelligent agents, not all relevant context is expressed in conversation. Thus, in this paper, we propose and analyze an algorithm, called Llumi, that incorporates other forms of context to learn to speak in long-term relationships modeled as …
Physics-Guided Machine Learning In Ocean Acoustics Using Fisher Information, Michael Craig Mortenson
Physics-Guided Machine Learning In Ocean Acoustics Using Fisher Information, Michael Craig Mortenson
Theses and Dissertations
Waterborne acoustic signals carry information about the ocean environment. Ocean geoacoustic inversion is the task of estimating environmental parameters from received acoustic signals by matching the measured sound with the predictions of a physics-based model. A lower bound on the uncertainty associated with environmental parameter estimates, the Cramér-Rao bound, can be calculated from the Fisher information, which is dependent on derivatives of a physics-based model. Physics-based preconditioners circumvent the need for variable step sizes when computing numerical derivatives. This work explores the feasibility of using a neural network to perform geoacoustic inversion for environmental parameters and their associated uncertainties from …
Deep Parameter Selection For Classic Computer Vision Applications, Michael Whitney
Deep Parameter Selection For Classic Computer Vision Applications, Michael Whitney
Theses and Dissertations
A trend in computer vision today is to retire older, so-called "classic'' methods in favor of ones based on deep neural networks. This has led to tremendous improvements in many areas, but for some problems deep neural solutions may not yet exist or be of practical application. For this and other reasons, classic methods are still widely used in a variety of applications. This paper explores the possibility of using deep neural networks to improve these older methods instead of replace them. In particular, it addresses the issue of parameter selection in these algorithms by using a neural network to …
Leveraging The Inductive Bias Of Large Language Models For Abstract Textual Reasoning, Christopher Michael Rytting
Leveraging The Inductive Bias Of Large Language Models For Abstract Textual Reasoning, Christopher Michael Rytting
Theses and Dissertations
Large natural language models (such as GPT-2 or T5) demonstrate impressive abilities across a range of general NLP tasks. Here, we show that the knowledge embedded in such models provides a useful inductive bias, not just on traditional NLP tasks, but also in the nontraditional task of training a symbolic reasoning engine. We observe that these engines learn quickly and generalize in a natural way that reflects human intuition. For example, training such a system to model block-stacking might naturally generalize to stacking other types of objects because of structure in the real world that has been partially captured by …
Methods For Generative Adversarial Output Enhancement, Michael B. Brodie
Methods For Generative Adversarial Output Enhancement, Michael B. Brodie
Theses and Dissertations
Generative Adversarial Networks (GAN) learn to synthesize novel samples for a given data distribution. While GANs can train on diverse data of various modalities, the most successful use cases to date apply GANs to computer vision tasks. Despite significant advances in training algorithms and network architectures, GANs still struggle to consistently generate high-quality outputs after training. We present a series of papers that improve GAN output inference qualitatively and quantitatively. The first chapter, Alpha Model Domination, addresses a related subfield of Multiple Choice Learning, which -- like GANs -- aims to generate diverse sets of outputs. The next chapter, CoachGAN, …
Trace: A Differentiable Approach To Line-Level Stroke Recovery For Offline Handwritten Text, Taylor Neil Archibald
Trace: A Differentiable Approach To Line-Level Stroke Recovery For Offline Handwritten Text, Taylor Neil Archibald
Theses and Dissertations
Stroke order and velocity are helpful features in the fields of signature verification, handwriting recognition, and handwriting synthesis. Recovering these features from offline handwritten text is a challenging and well-studied problem. We propose a new model called TRACE (Trajectory Recovery by an Adaptively-trained Convolutional Encoder). TRACE is a differentiable approach using a convolutional recurrent neural network (CRNN) to infer temporal stroke information from long lines of offline handwritten text with many characters. TRACE is perhaps the first system to be trained end-to-end on entire lines of text of arbitrary width and does not require the use of dynamic exemplars. Moreover, …
Semantic-Driven Unsupervised Image-To-Image Translation For Distinct Image Domains, Wesley Ackerman
Semantic-Driven Unsupervised Image-To-Image Translation For Distinct Image Domains, Wesley Ackerman
Theses and Dissertations
We expand the scope of image-to-image translation to include more distinct image domains, where the image sets have analogous structures, but may not share object types between them. Semantic-Driven Unsupervised Image-to-Image Translation for Distinct Image Domains (SUNIT) is built to more successfully translate images in this setting, where content from one domain is not found in the other. Our method trains an image translation model by learning encodings for semantic segmentations of images. These segmentations are translated between image domains to learn meaningful mappings between the structures in the two domains. The translated segmentations are then used as the basis …
Deep Learning To Predict Ocean Seabed Type And Source Parameters, David Franklin Van Komen
Deep Learning To Predict Ocean Seabed Type And Source Parameters, David Franklin Van Komen
Theses and Dissertations
In the ocean, light from the surface dissipates quickly leaving sound the only way to see at a distance. Different sediment types on the ocean floor and water properties like salinity, temperature, and ocean depth all change how sound travels across long distances. Hard sediment types, such as sand and bedrock, are highly reflective while softer sediment types, such as mud, are more absorptive and change the received sound upon arrival. Unfortunately, the vast majority of the ocean floor is not mapped and the expenses involved in creating such a map are far too great. Traditional signal processing methods in …
Facing The Hard Problems In Fgvc, Connor Stanley Anderson
Facing The Hard Problems In Fgvc, Connor Stanley Anderson
Theses and Dissertations
In fine-grained visual categorization (FGVC), there is a near-singular focus in pursuit of attaining state-of-the-art (SOTA) accuracy. This work carefully analyzes the performance of recent SOTA methods, quantitatively, but more importantly, qualitatively. We show that these models universally struggle with certain "hard" images, while also making complementary mistakes. We underscore the importance of such analysis, and demonstrate that combining complementary models can improve accuracy on the popular CUB-200 dataset by over 5%. In addition to detailed analysis and characterization of the errors made by these SOTA methods, we provide a clear set of recommended directions for future FGVC researchers.
Chaotic Model Prediction With Machine Learning, Yajing Zhao
Chaotic Model Prediction With Machine Learning, Yajing Zhao
Theses and Dissertations
Chaos theory is a branch of modern mathematics concerning the non-linear dynamic systems that are highly sensitive to their initial states. It has extensive real-world applications, such as weather forecasting and stock market prediction. The Lorenz system, defined by three ordinary differential equations (ODEs), is one of the simplest and most popular chaotic models. Historically research has focused on understanding the Lorenz system's mathematical characteristics and dynamical evolution including the inherent chaotic features it possesses. In this thesis, we take a data-driven approach and propose the task of predicting future states of the chaotic system from limited observations. We explore …
Deep Learning For Document Image Analysis, Christopher Alan Tensmeyer
Deep Learning For Document Image Analysis, Christopher Alan Tensmeyer
Theses and Dissertations
Automatic machine understanding of documents from image inputs enables many applications in modern document workflows, digital archives of historical documents, and general machine intelligence, among others. Together, the techniques for understanding document images comprise the field of Document Image Analysis (DIA). Within DIA, the research community has identified several sub-problems, such as page segmentation and Optical Character Recognition (OCR). As the field has matured, there has been a trend of moving away from heuristic-based methods, designed for particular tasks and domains of documents, and moving towards machine learning methods that learn to solve tasks from examples of input/output pairs. Within …
Using Perceptually Grounded Semantic Models To Autonomously Convey Meaning Through Visual Art, Derrall L. Heath
Using Perceptually Grounded Semantic Models To Autonomously Convey Meaning Through Visual Art, Derrall L. Heath
Theses and Dissertations
Developing advanced semantic models is important in building computational systems that can not only understand language but also convey ideas and concepts to others. Semantic models can allow a creative image-producing-agent to autonomously produce artifacts that communicate an intended meaning. This notion of communicating meaning through art is often considered a necessary part of eliciting an aesthetic experience in the viewer and can thus enhance the (perceived) creativity of the agent. Computational creativity, a subfield of artificial intelligence, deals with designing computational systems and algorithms that either automatically create original and functional products, or that augment the ability of humans …