Physical Sciences and Mathematics Commons^™

Interactive Contrastive Learning For Self-Supervised Entity Alignment, Kaisheng Zeng, Zhenhao Dong, Lei Hou, Yixin Cao, Minghao Hu, Jifan Yu, Xin Lv, Lei Cao, Xin Wang, Haozhuang Liu, Yi Huang, Jing Wan, Juanzi Li

Research Collection School Of Computing and Information Systems

Self-supervised entity alignment (EA) aims to link equivalent entities across different knowledge graphs (KGs) without the use of pre-aligned entity pairs. The current state-of-the-art (SOTA) selfsupervised EA approach draws inspiration from contrastive learning, originally designed in computer vision based on instance discrimination and contrastive loss, and suffers from two shortcomings. Firstly, it puts unidirectional emphasis on pushing sampled negative entities far away rather than pulling positively aligned pairs close, as is done in the well-established supervised EA. Secondly, it advocates the minimum information requirement for self-supervised EA, while we argue that self-described KG’s side information (e.g., entity name, relation name, …

Stitching Weight-Shared Deep Neural Networks For Efficient Multitask Inference On Gpu, Zeyu Wang, Xiaoxi He, Zimu Zhou, Xu Wang, Qiang Ma, Xin Miao, Zhuo Liu, Lothar Thiele, Zheng. Yang

Research Collection School Of Computing and Information Systems

Intelligent personal and home applications demand multiple deep neural networks (DNNs) running on resourceconstrained platforms for compound inference tasks, known as multitask inference. To fit multiple DNNs into low-resource devices, emerging techniques resort to weight sharing among DNNs to reduce their storage. However, such reduction in storage fails to translate into efficient execution on common accelerators such as GPUs. Most DNN graph rewriters are blind for multiDNN optimization, while GPU vendors provide inefficient APIs for parallel multi-DNN execution at runtime. A few prior graph rewriters suggest cross-model graph fusion for low-latency multiDNN execution. Yet they request duplication of the shared …

Application Of Nanodisc Technology To A Membrane-Bound P450: Functional Studies And Resonance Raman Studies On Cyp51a1, Yuanqi Jing

Dissertations (1934 -)

Membrane-bound cytochrome P450s are highly insoluble and tend to form aggregations in aqueous solution due to its hydrophobic N-terminal transmembrane anchor. The traditional methodology of studying membrane-bound proteins depends on the use of detergent; however, detergent solubilization of membrane P450s does not satisfy many criteria of biochemical and biophysical studies on membrane proteins. A well-developed Nanodisc technology has been proven to be an extraordinarily efficient method of studying membrane-bound P450s. The human sterol 14-demethylases (CYP51A1) are membrane-bound P450 enzymes that catalyze oxidative removal of the C32 methyl group of lanosterol by first forming an alcohol, then an aldehyde, and finally …

Kinetics Of Isotope-Labeled Pathways In The Ozone Forming Recombination Reaction, Elizaveta Grushnikova

Dissertations (1934 -)

The ozone layer in Earth’s atmosphere is unique and plays a vital role in the development of life. Studying the mechanism behind ozone formation helps us understand the development of our planet’s atmosphere. We focus here on the anomalous mass-independent isotope effect.1 In 1981, Mauersberger et al.2 performed an experiment using weather balloons, resulting in the discovery of the anomalous isotope effect for ozone formation. Since then, other chemists have continued to investigate the theory behind this phenomenon. To understand the nature of the isotope effect, we must consider all stages of ozone formation. The basic reaction for ozone formation …

Predicting Mental Health Crisis In Veterans: Early Warning Signs, Precursors And Protective Factors, Priyanka Annapureddy

Dissertations (1934 -)

Mental Health (MH) conditions have recently increased to a large extent due to socio-demographic changes. Posttraumatic Stress Disorder (PTSD) is one of the most common mental health disorders prevalent in US. PTSD is even more troubling at double the rate in combat veterans leaving their service compared to general population. Severity of PTSD is associated with risk taking behaviors such as substance abuse, non-suicidal self-injury, and sexual risk behaviors. Psychological disorders are often preceded by early warning signs and recognizing the early warning signs of PTSD will help in preventing the returning or worsening of PTSD symptoms. Ecological momentary assessment …

Bit-Flip Aware Data Structures For Phase Change Memory, Arockia David Roy Kulandai

Dissertations (1934 -)

Big, non-volatile, byte-addressable, low-cost, and fast non-volatile memories like Phase Change Memory are appearing in the marketplace. They have the capability to unify both memory and storage and allow us to rethink the present memory hierarchy. An important draw-back to Phase Change Memory is limited write-endurance. In addition, Phase Change Memory shares with other Non-Volatile Random Access Memories an asym- metry in the energy costs of writes and reads. Best use of Non-Volatile Random Access Memories limits the number of times a Non-Volatile Random Access Memory cell changes contents, called a bit-flip. While the future of main memory is still …

Adding Context To Source Code Representations For Deep Learning, Fuwei Tian, Christoph Treude

Research Collection School Of Computing and Information Systems

Deep learning models have been successfully applied to a variety of software engineering tasks, such as code classification, summarisation, and bug and vulnerability detection. In order to apply deep learning to these tasks, source code needs to be represented in a format that is suitable for input into the deep learning model. Most approaches to representing source code, such as tokens, abstract syntax trees (ASTs), data flow graphs (DFGs), and control flow graphs (CFGs) only focus on the code itself and do not take into account additional context that could be useful for deep learning models. In this paper, we …

Taming Multi-Output Recommenders For Software Engineering, Christoph Treude

Research Collection School Of Computing and Information Systems

Recommender systems are a valuable tool for software engineers. For example, they can provide developers with a ranked list of files likely to contain a bug, or multiple auto-complete suggestions for a given method stub. However, the way these recommender systems interact with developers is often rudimentary—a long list of recommendations only ranked by the model’s confidence. In this vision paper, we lay out our research agenda for re-imagining how recommender systems for software engineering communicate their insights to developers. When issuing recommendations, our aim is to recommend diverse rather than redundant solutions and present them in ways that highlight …

Video Graph Transformer For Video Question Answering, Junbin Xiao, Pan Zhou, Tat-Seng Chua, Shuicheng Yan

Research Collection School Of Computing and Information Systems

This paper proposes a Video Graph Transformer (VGT) model for Video Quetion Answering (VideoQA). VGT’s uniqueness are two-fold: 1) it designs a dynamic graph transformer module which encodes video by explicitly capturing the visual objects, their relations, and dynamics for complex spatio-temporal reasoning; and 2) it exploits disentangled video and text Transformers for relevance comparison between the video and text to perform QA, instead of entangled crossmodal Transformer for answer classification. Vision-text communication is done by additional cross-modal interaction modules. With more reasonable video encoding and QA solution, we show that VGT can achieve much better performances on VideoQA tasks …

Mix-Dann And Dynamic-Modal-Distillation For Video Domain Adaptation, Yuehao Yin, Bin Zhu, Jingjing Chen, Lechao Cheng, Yu-Gang Jiang

Research Collection School Of Computing and Information Systems

Video domain adaptation is non-trivial due to video is inherently involved with multi-dimensional and multi-modal information. Existing works mainly adopt adversarial learning and self-supervised tasks to align features. Nevertheless, the explicit interaction between source and target in the temporal dimension, as well as the adaptation between modalities, are unexploited. In this paper, we propose Mix-Domain-Adversarial Neural Network and Dynamic-Modal-Distillation (MD-DMD), a novel multi-modal adversarial learning framework for unsupervised video domain adaptation. Our approach incorporates the temporal information between source and target domains, as well as the diversity of adaptability between modalities. On the one hand, for every single modality, we …

Final Bpsou Subdrain Pump Station Remedial Design (Rd) Work Plan, Pioneer Technical Services, Inc.

Silver Bow Creek/Butte Area Superfund Site

No abstract provided.

Could A More Limited Environmental Goods Agreement Resolve Continued Issues In Cities Compliance?, Andrew Coccoli

William & Mary Environmental Law and Policy Review

This Note will adopt an economic perspective while advocating for the most sustainable practices in natural resources management. It will first conduct a series of case studies of natural resources for which there is transnational or international demand, but unsustainable management. While CITES-listed resources will be considered with particular attention, unlisted and more conventional resources will also be considered. In its second part, this Note will consider mechanisms currently available in international trade to accommodate environmental progress, then will set forth various new measures the WTO could adopt to incentivize sustainable management of the Part I resources. Part III will …

Oxygen Vacancies In Lib3O5 Crystals And Their Role In Nonlinear Absorption, Brian C. Holloway, Christopher A. Lenyk, Timothy D. Gustafson, Nancy C. Giles

Faculty Publications

LiB₃O₅ (LBO) crystals are used to generate the second, third, and fourth harmonics of near-infrared solid-state lasers. At high power levels, the material’s performance is adversely affected by nonlinear absorption. We show that as-grown crystals contain oxygen and lithium vacancies. Transient absorption bands are formed when these intrinsic defects serve as traps for “free” electrons and holes created by x rays or by three- and four-photon absorption processes. Trapped electrons introduce a band near 300 nm and trapped holes produce bands in the 500-600 nm region. Electron paramagnetic resonance (EPR) is used to identify and characterize the …

Asia’S Waste Crisis, Havovi Joshi

Asian Management Insights

A festering issue.

Transcriptional Profile Of Cohesin Complex Mutations In The Background Of Npm1 And Runx1-Runx1t1 Aml, Jacob Tiegs

Master's Theses (2009 -)

Acute Myeloid Leukemia is a cancer of the blood, characterized by a heterogenous mixture of disease causing mutations. Mutations of the cohesin complex is a group of such mutations and occur alongside several other driving mutations in the development of Acute Myeloid Leukemia. This thesis specifically focuses on cohesin complex mutations in the context of concurrence with NPM1 mutation and the Core Binding Factor (CBF) mutation RUNX1-RUNX1T1 in three distinct components. The first two components involved Differential Expression Analysis (DEA) to identify significantly differentiated genes in each model, followed by Gene Set Enrichment Analysis (GSEA) to identify Gene Ontology (GO) …

Gray Counters For Non-Volatile Memories, Arockia David Roy Kulandai, John Rose, Thomas Schwarz

Computer Science Faculty Research and Publications

New technologies for non-volatile memories combine the speed and byte addressability of current memory technologies with the low cost, density, and non-volatility of current storage technologies. They use energy only when writing or reading data. While some newer technologies have practically unlimited endurance, others, such as Phase Change Memory do not. However, this limited endurance surpasses that of solid state drives by several orders of magnitude. They can be integrated into the current memory storage hierarchy as a replacement for DRAM. To manage limited endurance, age-based wear leveling divides the memory into pages and counts the number of writes to …

Human Activity Recognition (Har) Using Wearable Sensors And Machine Learning, Chrisogonas Odero Odhiambo

Theses and Dissertations

Humans engage in a wide range of simple and complex activities. Human Activity Recognition (HAR) is typically a classification problem in computer vision and pattern recognition, to recognize various human activities. Recent technological advancements, the miniaturization of electronic devices, and the deployment of cheaper and faster data networks have propelled environments augmented with contextual and real-time information, such as smart homes and smart cities. These context-aware environments, alongside smart wearable sensors, have opened the door to numerous opportunities for adding value and personalized services to citizens. Vision-based and sensory-based HAR find diverse applications in healthcare, surveillance, sports, event analysis, Human-Computer …

Memristive Ising Circuits, Vincent James Dowling

Theses and Dissertations

Resistance random access memory (ReRAM) cells are emerging passive components showing great promise for applications in non-volatile storage, in-memory computing, and neuromorphic circuits, to name a few. Various experimental ReRAM cells have demonstrated excellent characteristics. This includes nanoscale size, short read and write times, and small energy requirements needed to manipulate the information. The main obstacles preventing the widespread adoption of ReRAM technology are significant device-to-device and cycle-to-cycle variability.

In essence, ReRAM cells are resistors with memory whose state (resistance) can be changed by an applied signal such as voltage. Many models were developed for describing their response. These range …

Public Acceptance Of Medical Screening Recommendations, Safety Risks, And Implied Liabilities Requirements For Space Flight Participation, Cory J. Trunkhill

Doctoral Dissertations and Master's Theses

The space tourism industry is preparing to send space flight participants on orbital and suborbital flights. Space flight participants are not professional astronauts and are not subject to the rules and guidelines covering space flight crewmembers. This research addresses public acceptance of current Federal Aviation Administration guidance and regulations as designated for civil participation in human space flight.

The research utilized an ordinal linear regression analysis of survey data to explore the public acceptance of the current medical screening recommended guidance and the regulations for safety risk and implied liability for space flight participation. Independent variables constituted participant demographic representations …

Conflict Dynamics In Scale-Free Networks With Degree Correlations And Hierarchical Structure, Eduardo Jacobo-Villegas, Bibiana Obregón-Quintana, Lev Guzmán-Vargas, Larry S. Liebovitch

Publications and Research

We present a study of the dynamic interactions between actors located on complex networks with scale-free and hierarchical scale-free topologies with assortative mixing, that is, correlations between the degree distributions of the actors. The actor’s state evolves according to a model that considers its previous state, the inertia to change, and the influence of its neighborhood. We show that the time evolution of the system depends on the percentage of cooperative or competitive
interactions. For scale-free networks, we find that the dispersion between actors is higher when all interactions are either cooperative or competitive, while a balanced presence of interactions …

Converting Hydrazone-Based Photoswitches Into Functional Materials, Sirun Yang

Dartmouth College Ph.D Dissertations

The development of new classes of molecular switches with enhanced performance and brand-new functionalities enables practitioners to push the frontiers of adaptive materials. In the realm of photoswitches, the bistability of hydrazones (i.e., molecules that incorporate the C=N-NH functional group) allows for the kinetic trapping of polymer and supramolecular assemblies resulting in multistate actuation and emergent phenomena that are not tractable with other photoswitches. This property has transformed hydrazone photoswitches into powerful tools that can be used in understanding fundamental molecular interactions and how to apply them in modulating the physicochemical properties of materials.

Progressing from disordered to …

Dynamics Of Atmospheric Gravity Wave Breaking In The Mesopause Region, Fan Yang

Doctoral Dissertations and Master's Theses

GWs significantly impact the Mesosphere and Lower Thermosphere (MLT), and as a consequence of GW breaking, atmospheric turbulence plays an essential role in the mixing and transport of momentum, mass, and chemical materials in MLT. The main focus of this research for the fulfillment of the proposed Ph.D. dissertation is to improve the understanding of the dynamic process of wave breaking and their effects in the mesopause region above Andes Lidar Observatory (30.2◦ S, 70.7◦ W). A few scientific topics related to GW break- ing are addressed. What are the probabilities of the atmosphere becoming convectively or dynamically unstable in …

Liposome Formulation For Tumor-Targeted Drug Delivery Using Radiation Therapy, Daniel Fologea

Physics Faculty Publications and Presentations

Targeted delivery of drugs or other therapeutic agents through internal or external triggers has been used to control and accelerate the release from liposomal carriers in a number of studies, but relatively few utilize energy of therapeutic X-rays as a trigger. We have synthesized liposomes that are triggered by ionizing radiation (RTLs) to release their therapeutic payload. These liposomes are composed of natural egg phosphatidylethanolamine (PE), 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC), cholesterol, and 1,2-disteroyl-sn-glycero-3-phosphoethanolamine-N-[methoxy (polyethylene glycol)-2000] (DSPE-PEG-2000), and the mean size of the RTL was in the range of 114 to 133 nm, as measured by nanoparticle tracking analysis (NTA). The trigger mechanism …

A Data Driven Modeling Approach For Store Distributed Load And Trajectory Prediction, Nicholas Peters

Doctoral Dissertations and Master's Theses

The task of achieving successful store separation from aircraft and spacecraft has historically been and continues to be, a critical issue for the aerospace industry. Whether it be from store-on-store wake interactions, store-parent body interactions or free stream turbulence, a failed case of store separation poses a serious risk to aircraft operators. Cases of failed store separation do not simply imply missing an intended target, but also bring the risk of collision with, and destruction of, the parent body vehicle. Given this risk, numerous well-tested procedures have been developed to help analyze store separation within the safe confines of wind …

Supporting The Discovery, Reuse, And Validation Of Cybersecurity Requirements At The Early Stages Of The Software Development Lifecycle, Jessica Antonia Steinmann

Doctoral Dissertations and Master's Theses

The focus of this research is to develop an approach that enhances the elicitation and specification of reusable cybersecurity requirements. Cybersecurity has become a global concern as cyber-attacks are projected to cost damages totaling more than $10.5 trillion dollars by 2025. Cybersecurity requirements are more challenging to elicit than other requirements because they are nonfunctional requirements that requires cybersecurity expertise and knowledge of the proposed system. The goal of this research is to generate cybersecurity requirements based on knowledge acquired from requirements elicitation and analysis activities, to provide cybersecurity specifications without requiring the specialized knowledge of a cybersecurity expert, and …

On Mitigating Hard Clusters For Face Clustering, Yingjie Chen, Huasong Zhong, Chong Chen, Chen Shen, Jianqiang Huang, Tao Wang, Yun Liang, Qianru Sun

Research Collection School Of Computing and Information Systems

Face clustering is a promising way to scale up face recognition systems using large-scale unlabeled face images. It remains challenging to identify small or sparse face image clusters that we call hard clusters, which is caused by the heterogeneity, i.e., high variations in size and sparsity, of the clusters. Consequently, the conventional way of using a uniform threshold (to identify clusters) often leads to a terrible misclassification for the samples that should belong to hard clusters. We tackle this problem by leveraging the neighborhood information of samples and inferring the cluster memberships (of samples) in a probabilistic way. We introduce …

Ngram-Oaxe: Phrase-Based Order-Agnostic Cross Entropy For Non-Autoregressive Machine Translation, Cunxiao Du, Zhaopeng Tu, Longyue Wang, Jing Jiang

Research Collection School Of Computing and Information Systems

Recently, a new training oaxe loss has proven effective to ameliorate the effect of multimodality for non-autoregressive translation (NAT), which removes the penalty of word order errors in the standard cross-entropy loss. Starting from the intuition that reordering generally occurs between phrases, we extend oaxe by only allowing reordering between ngram phrases and still requiring a strict match of word order within the phrases. Extensive experiments on NAT benchmarks across language pairs and data scales demonstrate the effectiveness and universality of our approach. Further analyses show that ngram noaxe indeed improves the translation of ngram phrases, and produces more fluent …

Autoprtitle: A Tool For Automatic Pull Request Title Generation, Ivana Clairine Irsan, Ting Zhang, Ferdian Thung, David Lo, Lingxiao Jiang

Research Collection School Of Computing and Information Systems

With the rise of the pull request mechanism in software development, the quality of pull requests has gained more attention. Prior works focus on improving the quality of pull request descriptions and several approaches have been proposed to automatically generate pull request descriptions. As an essential component of a pull request, pull request titles have not received a similar level of attention. To further facilitate automation in software development and to help developers draft high-quality pull request titles, we introduce AutoPRTitle. AutoPRTitle is specifically designed to generate pull request titles automatically. AutoPRTitle can generate a precise and succinct pull request …

Answer Summarization For Technical Queries: Benchmark And New Approach, Chengran Yang, Bowen Xu, Ferdian Thung, Yucen Shi, Ting Zhang, Zhou Yang, Xin Zhou, Jieke Shi, Junda He, Donggyun Han, David Lo

Research Collection School Of Computing and Information Systems

Prior studies have demonstrated that approaches to generate an answer summary for a given technical query in Software Question and Answer (SQA) sites are desired. We find that existing approaches are assessed solely through user studies. Hence, a new user study needs to be performed every time a new approach is introduced; this is time-consuming, slows down the development of the new approach, and results from different user studies may not be comparable to each other. There is a need for a benchmark with ground truth summaries as a complement assessment through user studies. Unfortunately, such a benchmark is non-existent …

Orb-Weaver Diversity And Niche Partitioning In Ecuador’S Amazonian Foothills: What Spiders Can Reveal About Tree Fall Gaps, Streams, And Cultivated Areas, Riley Endries

Independent Study Project (ISP) Collection

As abundant predators at the top of arthropod food chains, spiders are excellent bioindicators. Araneae is a megadiverse and extremely understudied order, especially in the tropics. This study aims to investigate the diversity of orb-weaver spiders across a disturbance gradient and variety of habitat types as well as their microhabitat preferences and potential niche partitioning. Spider collection was performed on spiders of the families Araneidae, Tetragnathidae, and Theridiosomatidae in Sumak Kawsay in Situ biological reserve in the lower elevation cloud forest of Ecuador’s Andean foothills. Spiders and webs were observed and analyzed from primary and secondary forest tree fall gaps, …