Digital Commons Network^™

Learning To See With Minimal Human Supervision, Zezhou Cheng

Doctoral Dissertations

Deep learning has significantly advanced computer vision in the past decade, paving the way for practical applications such as facial recognition and autonomous driving. However, current techniques depend heavily on human supervision, limiting their broader deployment. This dissertation tackles this problem by introducing algorithms and theories to minimize human supervision in three key areas: data, annotations, and neural network architectures, in the context of various visual understanding tasks such as object detection, image restoration, and 3D generation. First, we present self-supervised learning algorithms to handle in-the-wild images and videos that traditionally require time-consuming manual curation and labeling. We demonstrate that …

Monitoring Mammals At Multiple Scales: Case Studies From Carnivore Communities, Kadambari Devarajan

Doctoral Dissertations

Carnivores are distributed widely and threatened by habitat loss, poaching, climate change, and disease. They are considered integral to ecosystem function through their direct and indirect interactions with species at different trophic levels. Given the importance of carnivores, it is of high conservation priority to understand the processes driving carnivore assemblages in different systems. It is thus essential to determine the abiotic and biotic drivers of carnivore community composition at different spatial scales and address the following questions: (i) What factors influence carnivore community composition and diversity? (ii) How do the factors influencing carnivore communities vary across spatial and temporal …

Learning With Aggregate Data, Tao Sun

Doctoral Dissertations

Various real-world applications involve directly dealing with aggregate data. In this work, we study Learning with Aggregate Data from several perspectives and try to address their combinatorial challenges. At first, we study the problem of learning in Collective Graphical Models (CGMs), where only noisy aggregate observations are available. Inference in CGMs is NP- hard and we proposed an approximate inference algorithm. By solving the inference problems, we are empowered to build large-scale bird migration models, and models for human mobility under the differential privacy setting. Secondly, we consider problems given bags of instances and bag-level aggregate supervisions. Specifically, we study …

Bacterial Diversity And Function Within An Epigenic Cave System And Implications For Other Limestone Cave Systems, Kathleen Merritt Brannen-Donnelly

Doctoral Dissertations

There are approximately 48,000 known cave systems in the United States of America, with caves formed in carbonate karst terrains being the most common. Epigenic systems develop from the downward flow of meteoric water through carbonate bedrock and the solutional enlargement of interconnected subsurface conduits. Despite carbonate karst aquifers being globally extensive and important drinking water sources, microbial diversity and function are poorly understood compared to other Earth environments. After several decades of research, studies have shown that microorganisms in caves affect water quality, rates of carbonate dissolution and precipitation, and ecosystem nutrition through organic matter cycling. However, limited prior …

Geographic Disparities Associated With Stroke And Myocardial Infarction In East Tennessee, Ashley Pedigo Golden

Doctoral Dissertations

Stroke and myocardial infarction (MI) are serious conditions whose burdens vary by socio-demographic and geographic factors. Although several studies have investigated and identified disparities in burdens of these conditions at the county and state levels, little is known regarding their geographic epidemiology at the neighborhood level. Both conditions require emergency treatments and therefore timely geographic accessibility to appropriate care is critical. Investigation of disparities in geographic accessibility to stroke and MI care and the role of Emergency Medical Services (EMS) in reducing treatment delays are vital in improving health outcomes. Therefore, the objectives of this work were to: (i) classify …

Treeline Responses To Climate Change In High-Elevation Landscapes Of Western Montana, U.S.A., David F. Mann

Doctoral Dissertations

The purpose of this research was to use whitebark pine trees at three major sites in western Montana to: (1) determine tree species response to climate, (2) reconstruct past climate conditions, (3) determine the effects of climate shifts on treeline, and (4) reconstruct fire history from fire-scar data. I collected samples from whitebark pine and subalpine fir and from remnant whitebark pine in the western Beaverhead-Deerlodge and Lolo National Forests.

In the climate response analysis, the Palmer Drought Severity Index (PDSI) had the highest correlations with whitebark pine growth. The strongest relationship occurred in the previous year’s June and July. …

Parallel Simulation Of Individual-Based, Physiologically-Structured Population And Predator-Prey Ecology Models, Jeffrey A. Nichols

Doctoral Dissertations

Utilizing as testbeds physiologically-structured, individual-based models for fish and Daphnia populations, techniques for the parallelization of the simulation are developed and analyzed. The techniques developed are generally applicable to individual-based models. For rapidly reproducing populations like Daphnia which are load balanced, then global birth combining is required. Super-scalar speedup was observed in simulations on multi-core desktop computers.

The two populations are combined via a size-structured predation module into a predator-prey system with sharing of resource weighted by relative mass. The individual-based structure requires multiple stages to complete predation.

Two different styles of parallelization are presented. The first distributes both populations. …

Cenozoic And Recent Lunulitiform Bryozoans Of The Gulf And Atlantic Coasts, Ronald Greeley

Doctoral Dissertations

"Free, cup-shaped colonies of cheilostomatous bryozoans, equipped with vibracula, are termed lunulitiform bryozoans. This group is found in sandy and calcareous Cenozoic deposits of the Gulf and Atlantic Coastal Plains. Cupuliform bryozoans are taxonomically distinct from, but in many ways similar to, lunulitiform bryozoans; the two groups are frequently found together.

Known lunulitiform genera of the Gulf and Atlantic Coastal Plains are diagnosed and their species described: Lunulites Lamarck, fifteen species, one new; Oligotresium Gabb and Horn, six species, two new; Discoporella d'Orbigny, two species; Cupuladria Canu and Bassler, two species; Selenaria Busk, one species; Otionella Canu and Bassler, four …