Physical Sciences and Mathematics Commons^™

Examining Differences In Concept Representation Across Similarity Spaces Between Humans And Large Language Models, Krishnachandra Nair

Computer Science Senior Theses

The replication of human concept representation is a critical task for the pursuit of artificial general intelligence. With the recent influx of large language models that demonstrate text-generation capabilities nearly on par with humans, the question stands on whether these large language models can capture concepts within language. We examine this question by exploring differences in concept representation across similarity spaces between humans and LLMs. We find that, while concept representation within LLMs does partially mimic human concept representation, LLMs are greatly limited by their dependence on semantic information and cannot therefore develop an understanding of human social code or …

Ripl: Recursive Inference For Policy Learning, Kunal Jha, Jeremy R. Manning, Alberto Quattrini Li

Computer Science Senior Theses

Humans excel at understanding the thoughts and intentions of others (theory of mind) and leverage this ability to learn and adapt in social environments. However, replicating this capability in artificial agents remains a challenge. This paper explores the gap between fast, efficient learning often achieved by Reinforcement Learning (RL) algorithms and the interpretability and adaptability desired in agents interacting with humans. We propose a novel approach that integrates an inference network within existing RL frameworks. This allows agents to reason about the beliefs of others (nested reasoning) while learning optimal actions. Our method leverages approximate solutions to the I-POMDP framework, …

Dna-Templated Nanofabrication Of Metal-Semiconductor Heterojunctions And Their Electrical Characterization, Chao Pang

Theses and Dissertations

Bottom-up nanofabrication, although still in its early stages with formidable challenges, is considered a potential alternative method to address the limitations of traditional top-down techniques by offering benefits including process simplification, cost reduction, and environmental friendliness. DNA-templated nanofabrication, one of the most powerful bottom-up methods, presents an innovative way to create advanced nanoelectronics. In this approach, nanomaterials with specific electronic, photonic, or other functions are precisely and programmably positioned on DNA nanostructures from a disordered collection of smaller parts. These self-assembled structures offer significant potential for improving many fields such as biosensing, drug delivery and electronic device manufacturing. This dissertation …

Comparison Of Ligands In Palladium-Catalyzed Electrochemical Allyl 4-Pyridinylation, Wei-Jie Ding, Chun-Hui Yang, Zhong-Tao Feng, Shi-Rong Lu, Xu Cheng

Journal of Electrochemistry

4-CN-pyridine is a widely applied 4-pyridinylation reagent for diverse transformations. Conventionally, the reaction proceeds via an open-shell radical cross-coupling pathway. Following our previous study, in this work, we report the Pd-catalyzed allyl 4-pyrinylation reaction under electrochemical conditions. The reaction proceeds via radical-polar crossover pathway in which the role of phosphine ligand in reactivity and selectivity was extensively investigated.

Holographic Multiplexing In A Photopolymerisable Hybrid Sol-Gel, Pamela Stoeva, Tatsiana Mikulchyk, Izabela Naydenova, Kevin Murphy

Articles

Holographic multiplexing techniques enhance functionality and information storage by leveraging the inherent selectivity of holograms. This is crucial for advancing holographic sensors, which excel in simultaneously detecting multiple parameters from a single input signal. This study explores the potential of the recent photopolymerisable hybrid sol-gel (PHSG) material for application in Space sensing systems through the investigation of its holographic angular multiplexing capabilities. For the first time, to the best of our knowledge, we report the successful recording of up to five angularly multiplexed gratings with diffraction efficiencies (DE) ≥ 15% in 187 ± 18 µm PHSG layers. A 3 mW/cm2 …

Investigation Of Volatility, Composition, And Gas-Particle Phase Partitioning Of Atmospheric Organic Compounds Through Novel Instrumentation And Techniques, Karolina Cysneiros De Carvalho

McKelvey School of Engineering Graduate Student Theses & Dissertations

Atmospheric aerosols are ubiquitous indoors and outdoors and their impact on human life on Earth is extensive. Aerosol particles scatter and absorb solar radiation, are key in the formation of clouds and precipitation, and can affect the abundance and distribution of greenhouse and atmospheric trace gases by physicochemical multiphase processes, thus they play an important role in regulating regional and global climate. On the other hand, poor indoor and outdoor air quality associated with high particulate matter (PM) levels is among the leading health risks worldwide, affecting life quality and expectancy by increasing the risk of cancer, cardiovascular and respiratory …

Re: Silver Bow Creek Conservation Area Repository Data Gap Quality Assurance Project Plan (Qapp), Josh Bryson

Silver Bow Creek/Butte Area Superfund Site

No abstract provided.

Recent Advance In Electrochemical Dehalogenative Deuteration, Peng-Fei Li, Guang-Sheng Kou, Li-Ping Qi, You-Ai Qiu

Journal of Electrochemistry

In recent years, the incorporation of deuterium atoms into organic compounds has emerged as a vital focus in the development of pharmaceutical molecules. This trend is driven by the increasing recognition of the significance of compounds containing deuterium atoms across various domains, including materials and biopharmaceuticals, where they have found widespread applications in mechanistic studies within the realms of chemistry and biology. Meanwhile, organic electrochemistry, as a relatively environmentally friendly catalytic mode with broad adaptability to redox reactions, has emerged as a crucial alternative to traditional halogen-deuterium exchange in the context of the reduction deuteration of halides. This approach circumvents …

Nickel-Catalyzed Α-Arylation Of Α-Cyanoacetates Enabled By Electrochemistry, Zi-Meng Li, Zhang-Jian Li, Anat Milo, Ping Fang, Tian-Sheng Mei

Journal of Electrochemistry

β-Amino acids have a wide range of applications in the field of pharmaceuticals. Utilizing a combination strategy of nickel catalysis and paired electrolysis, a catalytic α-arylation protocol of carbonyl compounds has been developed. This protocol affords various a-aryl-a-cyanoacetates, which can be reduced to high-value-added α-aryl-β-amino acids. The cross-coupling reaction of electron-deficient aryl bromides with a-cyanoacetates achieves the expected products with good yields and functional group compatibility under mild conditions. Excessive electron-richness in initial aryl bromides facilitates the self-coupling of desired products. DFT calculations confirm that the presence of electron-rich aryl substitutions decreases the reduction potentials of the product anions, making …

Advancing Objective Mobile Device Use Measurement Inchildren Ages 6–11 Through Built-In Device Sensors: A Proof-Of-Concept Study, Olivia L. Finnegan, Robert Glenn Weaver Med, Phd, Hongpeng Yang, James W. White, Srihari Nelakuditi, Zifei Zhong, Rahul Ghosal Ph.D., Yan Tong, Aliye B. Cepni, Elizabeth L. Adams, Sarah Burkart Mph, Ph.D., Michael W. Beets Med, Mph, Phd, Bridget Armstrong Ph.D.

Faculty Publications

Mobile devices (e.g., tablets and smartphones) have been rapidly integrated into the lives of children and have impacted howchildren engage with digital media. The portability of these devices allows for sporadic, on-demand interaction, reducing theaccuracy of self-report estimates of mobile device use. Passive sensing applications objectively monitor time spent on a givendevice but are unable to identify who is using the device, a significant limitation in child screen time research. Behavioralbiometric authentication, using embedded mobile device sensors to continuously authenticate users, could be applied toaddress this limitation. This study examined the preliminary accuracy of machine learning models trained on iPad …

Entertainment-Education And Climate Change: Program Examples, Evidence, And Best Practices From Around The World, Suruchi Sood, Amy Henderson Riley, Lyena Birkenstock

College of Population Health Faculty Papers

Entertainment-education is a global theory-driven and evidence-based storytelling strategy that promotes social and behavioral change. A formal review of the peer-reviewed and grey literature, alongside consultations with program and industry experts, revealed several contemporary examples of entertainment-education and climate change. Late-night comedy, with the inclusion of climate change storylines within popular entertainment, is common in the Global North. In the Global South, climate adaptation and mitigation narratives are situated in formats portraying audiences’ lived experiences. Crosscutting trends relate to documentary storytelling, children’s programming, and future efforts. Recommendations include re-examining theories across the social-ecological model; placing entertainment front and center; including …

Novel In Situ Synthesis Of Copper Oxide Nanoparticles In Epoxy Network: Kinetics, Composite Mechanical And Dielectric Properties, Elena Bobina, Maxim Danilaev, Safaa.M.R.H. Hussein, Sergey Karandashov, Vladimir Kuklin, Ivan Lounev, Konstantin Faizullin

Karbala International Journal of Modern Science

>Mechanical properties of polymer composites with dispersed nanoparticles (CDNP) depend on interaction between the nanoparticles and the polymer matrix. Strength of polymer composites significantly decreases when there is no interaction between dispersed nanoparticles and the polymer. This limits the application of functional polymer composites with dispersed nanoparticles. In this study, CDNP based on ED-20 epoxy resin with dispersed copper oxide nanoparticles was obtained.These nanoparticles were synthesized in epoxy resin before curing: the nanoparticles were obtained by decomposition of copper hydroxide by heating its solution in ED-20 resin.The kinetics of copper oxide nanoparticles formation in CDNP samples were studied using two …

Elm And Lightgbm: A Hybrid Machine Learning Technique With Intelligent Iot To Predict The Cardiovascular Disease, Gorapalli Srinivasa Rao, G Muneeswari

Karbala International Journal of Modern Science

Cardiologists can more accurately classify a patient's condition by performing an accurate diagnostic and prognosis of cardiovascular disease (CVD). The clinical diagnosis, and therapies processes within the medical field have been substantially accelerated by ML-based approaches enabled by IoT-based systems. This structure is based on IoT-based system with enabled ML approach. This study investigates an approach known as ensemble categorization, which enhances the precision of weak algorithms by integrating multiple classifiers. For effective CVD classification, we utilized Ensemble learning machine (ELM) and Light GBM. The appropriate traits are chosen to speed up the categorization process using the Gorilla Troops Optimizer …

Comparing Researchers’ Degree Of Dichotomous Thinking Using Frequentist Versus Bayesian Null Hypothesis Testing, Jasmine Muradchanian, Rink Hoekstra, Henk Kiers, Dustin Fife, Don Van Ravenzwaaij

College of Science & Mathematics Departmental Research

A large amount of scientific literature in social and behavioural sciences bases their conclusions on one or more hypothesis tests. As such, it is important to obtain more knowledge about how researchers in social and behavioural sciences interpret quantities that result from hypothesis test metrics, such as p-values and Bayes factors. In the present study, we explored the relationship between obtained statistical evidence and the degree of belief or confidence that there is a positive effect in the population of interest. In particular, we were interested in the existence of a so-called cliff effect: A qualitative drop in the …

Enhanced Dip Angle Map Using Kuwahara And Gaussian Filters: An Example From Burdur Region, Türkiye, Özkan Kafadar, Erdi̇nç Öksüm

Turkish Journal of Earth Sciences

Recently, many edge detection and enhancement techniques have been used to transform geophysical data into a more suitable form for interpretation. The dip angle, a surface-derived attribute that has been frequently used in the interpretation of seismic data, is one of these transforms, and recently, it has started to be utilized as an edge detector in the interpretation of potential field data. In this study, it is proposed to use the potential field data filtered using the Kuwahara and Gaussian filters to obtain clearer and more detailed edges from both the conventional dip angle and its rescaled version. Two synthetic …

Analysis Of Green Data Center Efforts And Energy Usage, Dillon J. Goicoechea

Honors Projects

This paper is an undergraduate level literature review and analysis of research surrounding the Green Data Center phenomenon. Review of work covering energy usage, data usage, usage predictions, and strategies for decreasing energy requirements is the main analysis of this work. The analysis shows that while data centers are becoming greener, the increase in usage of their capacities is negating those efficiency increases. The increase in the energy efficiency of data centers is crucial, however, there must be made efforts to lower computational and data usage to help achieve lower energy usage of data centers.

Derivation Of A 1-D Seismic Velocity Model For The Lake Van Region, Mehveş Feyza Akkoyunlu, Bülent Kaypak, Bülent Oruç, Doğan Kalafat

Turkish Journal of Earth Sciences

We present a detailed one-dimensional (1-D) velocity model of the Lake Van Basin and its surrounding region, alongside precise hypocenter locations, utilizing seismic data from the aftershocks of the 23 October 2011 Lake Van earthquake (Mw = 7.1). The 1-D minimum velocity model is computed by the VELEST program by analyzing body wave traveltimes and station corrections. Encompassing a geographical area of 180 × 90 km² centered around Lake Van, the study incorporated seismic data from 10 newly deployed stations, complementing the existing network of the Kandilli Earthquake Research Institute. A comprehensive manual analysis was conducted on a dataset comprising …

Geomagnetic Response To The Earthquake In Türkiye And Syria On February 6, 2023, Svetlana Riabova, Sergey Shalimov

Turkish Journal of Earth Sciences

The disturbances in the lower ionosphere during the strong earthquake in Türkiye and Syria and its strongest aftershock on February 6, 2023, are analyzed using data from ground-based magnetometers. The observation points are located at distances 700 to 1600 km from the epicenter of the seismic event. As a characteristic of the ionospheric response to these events, variations in the magnetic field have been analyzed at the İznik, Grocka, Panagjurishte, and Surlari magnetic observatories. Earthquake epicenter is known to be a source of both seismic Rayleigh waves and atmospheric acoustic-gravity waves. Using the difference in velocities, at which these waves …

Stochastic 1-D Reactive Transport Simulations To Assess Silica And Carbonate Phases During The Co2 Reinjection Process In Metasediments, Selçuk Erol

Turkish Journal of Earth Sciences

One proposed method to mitigate carbon emission is to mineralize the CO2 in deep geothermal reservoirs while mixing the coproduced CO2 with the effluent fluid for reinjection. The injection fluid temperature fluctuates due to the mixing process between CO2-charged water and the effluent fluid, and compressor interruptions change the thermodynamic conditions that influence the fluid-rock interaction in the reservoir. Mineral dissolution or precipitations are associated with changes in permeability and porosity that affect the flow and, eventually, the lifespan of the reservoir. A combined stochastic–reactive transport simulation approach is useful for inspection purposes. Moreover, the stochastic algorithm validates the deterministic …

Carboniferous Graptolites (Hemichordata: Graptolithina) From The Dnipro-Donets Depression And Donets Basin, Ukraine, Vitaly Dernov

Turkish Journal of Earth Sciences

Remains of the Carboniferous graptolites ?Ptiograptus sp. and Dictyonema sp. are described from the Donets Basin and the border zone of the Dnipro-Donets Depression and Donets Basin (eastern Ukraine). These graptolites were found in the Mezhova (Visean, Mississippian) and Avilovka (Kasimovian, Pennsylvanian) formations. Dictyonema sp. from the Avilovka Formation is probably the youngest dendroid graptolite to date. The analysis of geological data allowed to make a reasonable assumption about the Late Devonian rather than the Carboniferous age of the graptolite fauna from the Englewood Formation of South Dakota (USA) and the Tournaisian rather than the Cisuralian age of graptolites from …

Deposition Of Organic And Inorganic Carbons On The Turkish Continental Margins (1984–1996), Vedat Edi̇ger, Di̇lek Edi̇ger

Turkish Journal of Earth Sciences

This study reevaluates the published total inorganic carbon (TIC) and total organic carbon (TOC) percentages of 695 seafloor sediment samples collected from the continental margins of the Black Sea, Sea of Marmara, Aegean Sea and Mediterranean Sea between 1984 and 1996. An inverse relationship is observed between the average TIC and TOC percentages in the four seas surrounding Türkiye. The explanation for this phenomenon is closely connected to the terrestrial, marine, climatic, and environmental factors of the continental margins from which the samples were collected.

Biotechnology For Sustainable Materials: Innovating Today For A Greener Tomorrow, Shashi Kant Bhatia, Saurabh Bhatia, Maria Eugenia Inda-Webb, Konstantina Kourmentza, Tae Seok Moon, Vijai Singh, Vishal Ahuja, Jingbo Li, Sanjeet Mehariya, Jikai Zhao

School of Earth, Environmental, & Marine Sciences Faculty Publications

No abstract provided.

Balanced N-Color Compositions, Jesus Omar Sistos Barron, Hua Wang

Mathematical Sciences: Faculty Publications

We introduce the notion of balanced n-color compositions, and consider the related enumeration problems. We also establish bijections with four other combinatorial objects, namely two types of Motzkin lattice paths, pairs of regular compositions with restricted parts, and regular compositions with a centered maximum.

On A Generalization Of A Theorem Of Ibukiyama To Evaluate Three Imprimitive Character Sums, Brad Isaacson

Publications and Research

In a previous paper, we expressed three families of character sums by certain generalized Bernoulli functions which in turn were expressed by generalized Bernoulli numbers via a complicated and indirect process. In this paper, we generalize a theorem of Ibukiyama to directly express these generalized Bernoulli functions by generalized Bernoulli numbers. As a result, we can express the three families of character sums by generalized Bernoulli numbers in a more elegant fashion than was done before.

A Novel Consumer-Centric Metric For Evaluating Hearing Device Audio Performance, Vinaya Manchaiah, Steve Taddei, Abram Bailey, De Wet Swanepoel, Hansapani Rodrigo

School of Mathematical & Statistical Sciences Faculty Publications

Background and Aim: The emergence of direct-to-consumer hearing devices has introduced confusion in making appropriate choices, highlighting the need for users to be well-informed for optimal device selection. Currently, no established metric offers insights into the sound performance of these devices. This study aimed to introduce and assess a novel consumer-centric metric (i.e., SoundScore) for hearing device audio performance.

Method: The SoundScore metric was created based on five dimensions of hearing device audio performance (i.e., speech benefit in quiet and moderate, speech benefit in loud, own voice perception, feedback control, streamed music sound quality). Tests were conducted under lab conditions …

A Combinatorial Proof Of A Partition Perimeter Inequality, Hunter Waldron

Michigan Tech Publications

The partition perimeter is a statistic defined to be one less than the sum of the number of parts and the largest part. Recently, Amdeberhan, Andrews, and Ballantine proved the following analog of Glaisher’s theorem: for all m ≥ 2 and n ≥ 1, there are at least as many partitions with perimeter n and parts repeating fewer than m times as there are partitions with perimeter n with parts not divisible by m. In this work, we provide a combinatorial proof of their theorem by relating the combinatorics of the partition perimeter to that of compositions. Using this technique, …

Composition-Theoretic Series And False Theta Functions, William Keith, Robert Schneider, Andrew V. Sills

Michigan Tech Publications

Many natural partition-theoretic series can be equally readily interpreted as compo-sition-theoretic series, but this viewpoint seems to have not been much employed in either theory. We consider some of the consequences of this viewpoint. As examples, we give results concerning the reciprocals of Ramanujan’s theta functions and of the false theta functions of L. J. Rogers, and raise an array of questions related to these. Part of this study may be considered a natural dual of the truncated pentagonal number theorem of Andrews and Merca.

Bagging Improves The Performance Of Deep Learning-Based Semantic Segmentation With Limited Labeled Images: A Case Study Of Crop Segmentation For High-Throughput Plant Phenotyping, Yinglun Zhan, Yuzhen Zhou, Geng Bai, Yufeng Ge

Department of Statistics: Faculty Publications

Advancements in imaging, computer vision, and automation have revolutionized various fields, including field-based high-throughput plant phenotyping (FHTPP). This integration allows for the rapid and accurate measurement of plant traits. Deep Convolutional Neural Networks (DCNNs) have emerged as a powerful tool in FHTPP, particularly in crop segmentation—identifying crops from the background—crucial for trait analysis. However, the effectiveness of DCNNs often hinges on the availability of large, labeled datasets, which poses a challenge due to the high cost of labeling. In this study, a deep learning with bagging approach is introduced to enhance crop segmentation using high-resolution RGB images, tested on the …

Carbon Sequestration In Peatlands: Using Environmental Proxies To Understand The Impact Of A Changing Climate On Global Carbon Storage, Sarah Schuller, Don Sullivan

DU Undergraduate Research Journal Archive

Peatlands are a type of terrestrial wetland ecosystem in which consistently waterlogged conditions prevent the decomposition of organic matter. These conditions allow for sequestered carbon in plant matter to remain stored in the soil to such a great degree that peatlands store more carbon than all other vegetation types in the world combined. By inhibiting decomposition, the composition of water-rich peat soil remains representative of the environmental conditions during the period in which the peat was formed. The collected peat samples can then be utilized as environmental proxies to determine historical temperature, moisture, and carbon content and extrapolated to predict …

Du Undergraduate Showcase Abstracts: Research, Scholarship, And Creative Works, August Alexander, Kevin Summers, E. Paige Lloyd, Alyssa Aragon, Lauren Wols, Erica Larson, Colton Arciniaga, Lily Baeza, Ellie Barnett-Cashman, Sidney Barbier, Maverick Bartholomew, Ryan Bell, Trevor Briggs, Amanda Brown, Cory Marchi, Maddy Pontius, Jennifer Gallagher, Claire Kwok, Channing Bullock, Alex Mccollister, Kim Gorgens, Daniela Chavez, Yamilet Espinoza Nuñez, Giovanni Valladares Giron, William Christensen, Olivia Kachulis, Daniel Cierasynski, Brandon Cohen, Spenser Dobbs, Lindsay Goolsby, Noah Craver, Nyah Cubbison, Liam Doherty, Nicole Doris, Yan Qin, Alexander Nguyen, Lily Harmon, Shubh Todi, Naichen Zhao, Noah Fagello, Stacia Fritz, Paola Gascot-Chinea, Alex Volkova, Janelyn Geronimo, Elsie Harrington, Shane Simmons, Clarice Hise, Laura Moreno Palmer, Henry Xu, Yesu Leela Pasupula, Emily Fisher, Zachary Hogan, Colin Kleckner, Alyssa Knaus, Erin Kubat, Sam Werkema, Katie Lamberton, Lilliya Larson, Allie Leary, Vanessa Leon-Gamez, James H. Gallagher, E. Dale Broder, Robin M. Tinghitella, Emma Loeber, Alina Mali, Ixchel Marquez, Lauren Mcgrath, Ella Matthews, Grace Naegelen, Audrey Ng, Megan Lucyshyn, Rachel Brough, Alexandra Norman, Joe Ontiveros Rodriguez, Ben Dossett, Ori Miller, Janamejay Sharma, Christopher Reardon, Vincent Pandey, Regan Parish, Ren Pratt, Alisha Pravasi, Sanchari Das, Cari Reichel, Emma Robson, Victoria Rockwell, Jayce Rumsey, Steven Said, Norah Schroder, Alaina Smith, Shannon Murphy, Shujan A. Sharafeldeen, Stanley M. Kanai, James T. Nichols, David E. Clouthier, Macalia R. Augustus, Daniel Silva Rios, Scott Simpson, Anna Sparling, Chase Spurbeck, Kimberly Chiew, Christine Stadnik-Poteroba, Jacqueline Stephenson, Evelyn Stovin, Mia Supan, Lauren Tapper, Will Thrush, Anna Vogt, Michaela Walheim, Jon Weber, Hunter Whitehouse, Kansas Wood, Hayley Sayre, Lydia Mccann, Maread Mclaughlin, Kelly Krumrie, Olivia Wuttke

DU Undergraduate Research Journal Archive

Abstracts from the DU Undergraduate Research Showcase.