Physical Sciences and Mathematics Commons^™

Hierarchical Multimodal Attention For End-To-End Audio-Visual Scene-Aware Dialogue Response Generation, Hung Le, Doyen Sahoo, Nancy F. Chen, Steven C. H. Hoi

Research Collection School Of Computing and Information Systems

This work is extended from our participation in the Dialogue System Technology Challenge (DSTC7), where we participated in the Audio Visual Scene-aware Dialogue System (AVSD) track. The AVSD track evaluates how dialogue systems understand video scenes and responds to users about the video visual and audio content. We propose a hierarchical attention approach on user queries, video caption, audio and visual features that contribute to improved evaluation results. We also apply a nonlinear feature fusion approach to combine the visual and audio features for better knowledge representation. Our proposed model shows superior performance in terms of both objective evaluation and …

Bus Frequency Optimization: When Waiting Time Matters In User Satisfaction, Songsong Mo, Zhifeng Bao, Baihua Zheng, Zhiyong Peng

Research Collection School Of Computing and Information Systems

Reorganizing bus frequency to cater for the actual travel demand can save the cost of the public transport system significantly. Many, if not all, existing studies formulate this as a bus frequency optimization problem which tries to minimize passengers’ average waiting time. However, many investigations have confirmed that the user satisfaction drops faster as the waiting time increases. Consequently, this paper studies the bus frequency optimization problem considering the user satisfaction. Specifically, for the first time to our best knowledge, we study how to schedule the buses such that the total number of passengers who could receive their bus services …

On Modeling Labor Markets For Fine-Grained Insights, Hendrik Santoso Sugiarto, Ee-Peng Lim

Research Collection School Of Computing and Information Systems

The labor market consists of job seekers looking for jobs, and job openings waiting for applications. Classical labor market models assume that salary is the primary factor explaining why job-seekers select certain jobs. In practice, job seeker behavior is much more complex and there are other factors that should be considered. In this paper, we therefore propose the Probabilistic Labor Model (PLM) which considers salary satisfaction, topic preference matching, and accessibility as important criteria for job seekers to decide when they apply for jobs. We also determine the user and job latent variables for each criterion and define a graphical …

Pricing And Equilibrium In On-Demand Ride-Pooling Markets, Jintao Ke, Hai Yang, Xinwei Li, Hai Wang, Jieping Ye

Research Collection School Of Computing and Information Systems

With the recent rapid growth of technology-enabled mobility services, ride-sourcing platforms, such as Uber and DiDi, have launched commercial on-demand ride-pooling programs that allow drivers to serve more than one passenger request in each ride. Without requiring the prearrangement of trip schedules, these programs match on-demand passenger requests with vehicles that have vacant seats. Ride-pooling programs are expected to offer benefits for both individual passengers in the form of cost savings and for society in the form of traffic alleviation and emission reduction. In addition to some exogenous variables and environments for ride-sourcing market, such as city size and population …

A Genetic Algorithm To Minimise Number Of Vehicles In An Electric Vehicle Routing Problem, Kiian Leong Bertran Queck, Hoong Chuin Lau

Research Collection School Of Computing and Information Systems

Electric Vehicles (EVs) and charging infrastructure are starting to become commonplace in major cities around the world. For logistics providers to adopt an EV fleet, there are many factors up for consideration, such as route planning for EVs with limited travel range as well as long-term planning of fleet size. In this paper, we present a genetic algorithm to perform route planning that minimises the number of vehicles required. Specifically, we discuss the challenges on the violations of constraints in the EV routing problem (EVRP) arising from applying genetic algorithm operators. To overcome the challenges, techniques specific to addressing the …

An Empirical Study Of The Dependency Networks Of Deep Learning Libraries, Junxiao Han, Shuiguang Deng, David Lo, Chen Zhi, Jianwei Yin, Xin Xia

Research Collection School Of Computing and Information Systems

Deep Learning techniques have been prevalent in various domains, and more and more open source projects in GitHub rely on deep learning libraries to implement their algorithms. To that end, they should always keep pace with the latest versions of deep learning libraries to make the best use of deep learning libraries. Aptly managing the versions of deep learning libraries can help projects avoid crashes or security issues caused by deep learning libraries. Unfortunately, very few studies have been done on the dependency networks of deep learning libraries. In this paper, we take the first step to perform an exploratory …

2020 Chesapeake Bay Dead Zone Report, Virginia Institute Of Marine Science, Anchor Qea

Reports

The “Dead Zone” of the Chesapeake Bay refers to a volume of bottom water that is characterized by dissolved oxygen concentrations less than 2 mg/L, which is too low for aquatic organisms such as fish and blue crabs to thrive.The Chesapeake Bay experiences such “hypoxic”conditions every year, with the severity varying from year to year, depending on nutrient and freshwater inputs, wind, and temperature. Multiple metrics are used to relate the severity of hypoxia between different years:

• Hypoxic Duration (days): The number of days in a given year between the first and last day of hypoxic conditions exceeding 2 km3in …

Seasonal Infilling And Sedimentary Characteristics In Sandy Versus Muddy Coastal Borrow Areas On The Louisiana Continental Shelf, Usa, Matthew Barley

LSU Master's Theses

Offshore sand deposits on the Louisiana Continental Shelf, such as inner shelf shoals and buried paleo-river channels, can be excavated to restore beaches and barrier islands that are rapidly deteriorating due to subsidence, sea-level rise and deficits in coastal sediment supply. Presented here is grain size, x-radiograph, and Beryllium-7 (⁷Be) derived sedimentation rates from multicores (~ 50 cm depth) retrieved from borrow areas (BAs) in contrasting depositional settings, all of which have implications for management of water quality, seafloor sedimentology, and biogeochemistry in proximal areas. Multicores were retrieved in fall 2018 at Caminada BA — a sandy energetic …

Hydrodynamics, Sediment Transport, And Water Quality Of Two Contrasting Dredge Pits On The Louisiana Shelf, Robert Bales

LSU Master's Theses

Sediment is needed for coastal restoration in Louisiana and is often excavated offshore from both mud-capped and sandy dredge pits. The mined sand from mud-capped pits, like Sandy Point which is west of modern Mississippi Delta of Louisiana, was originally from a paleo river channel and later covered by mud from modern coastal processes. Sandy pits, like Caminada pit in eastern Ship Shoal on Louisiana shelf, generally experience higher energy conditions. To better understand the post-dredging effects on hydrodynamics, sediment transport, and water quality, two tripods were deployed at Caminada and Sandy Point pits in summer 2018 and 2019, respectively, …

Effect Of Gradually Water Stress And Recovery On Photosynthesis, Transpiration, And Stomatal Conductance In Two Plantago Species, A. Rahimi, H. R. Rahimian Mashhadi, M. R. Jahansoz

IGC Proceedings (1993-2023)

No abstract provided.

Correlations Between Soil Properties And Plant Growth For Special Synthetic Soils Added With Five Components Used In High‐And‐Cut Rock Slopes, Guijuan Gao, Ruihong Han

IGC Proceedings (1993-2023)

No abstract provided.

Study On The Relationship Between Edaphon Amount And Soil Type In Desert Grassland, Ting Gao, Fengxia Li, Xiaogang Zhang

IGC Proceedings (1993-2023)

No abstract provided.

Nutrition Management In Fodder Grass Production For Fisheries In Hubei, F. Chen, J. W. Lu, X. K. Li, W. X. Li, H. X. Zhao, J. M. Lu

IGC Proceedings (1993-2023)

No abstract provided.

Tommy John Surgery: Potential Risk Factors And Causes In Major League Pitchers, Ethan Rhinehart

Analytics Capstones

Since 1974, over 270 Tommy John surgeries have been performed on pitchers at the major league level. Thousands more surgeries have been performed on minor league, college, high school and youth pitchers. As more biomechanical and statistical research has been conducted over the past few decades, a clearer picture of some of the risks and causes that lead to serious elbow injuries in pitchers have been found. This paper explores the research surrounding several of those factors, including pitching mechanics, pitch velocity, and pitch type. Using a data set comprised of major league pitchers that have undergone Tommy John surgery …

Impact Of Lost Gas Tax Revenue Due To Sale Of Electric Vehicles: Analysis And Recommendations For The 50 States, Jennifer Ricciuti

Analytics Capstones

Although states might have policy reasons to encourage the use of Electric Vehicles (EVs), the impact of future U.S. EV sales present a significant loss of gas tax revenue for each of the states, as these vehicles do not require gas to operate. For the last three years the number of Electric Vehicle registrations have doubled and are steadily increasing as a result of people becoming more economically and ecologically minded. This is proving to be an optimal choice for car purchasers over standard Internal Combustion Engine (ICE) vehicles, as research has shown that Electric Vehicles are superior for exhibiting …

Efficient Time-Stepping Approaches For The Dispersive Shallow Water Equations, Linwan Feng

Dissertations

This dissertation focuses on developing efficient and stable (high order) time-stepping strategies for the dispersive shallow water equations (DSWE) with variable bathymetry. The DSWE extends the regular shallow water equations to include dispersive effects. Dispersion is physically important and can maintain the shape of a wave that would otherwise form a shock in the shallow water system.

In some cases, the DSWE may be simplified when the bathymetry length scales are small (or large) in relation to other length scales in the shallow water system. These simplified DSWE models, which are related to the full DSWEs, are also considered in …

Hybrid Deep Neural Networks For Mining Heterogeneous Data, Xiurui Hou

Dissertations

In the era of big data, the rapidly growing flood of data represents an immense opportunity. New computational methods are desired to fully leverage the potential that exists within massive structured and unstructured data. However, decision-makers are often confronted with multiple diverse heterogeneous data sources. The heterogeneity includes different data types, different granularities, and different dimensions, posing a fundamental challenge in many applications. This dissertation focuses on designing hybrid deep neural networks for modeling various kinds of data heterogeneity.

The first part of this dissertation concerns modeling diverse data types, the first kind of data heterogeneity. Specifically, image data and …

Enrichment Of Ontologies Using Machine Learning And Summarization, Hao Liu

Dissertations

Biomedical ontologies are structured knowledge systems in biomedicine. They play a major role in enabling precise communications in support of healthcare applications, e.g., Electronic Healthcare Records (EHR) systems. Biomedical ontologies are used in many different contexts to facilitate information and knowledge management. The most widely used clinical ontology is the SNOMED CT. Placing a new concept into its proper position in an ontology is a fundamental task in its lifecycle of curation and enrichment.

A large biomedical ontology, which typically consists of many tens of thousands of concepts and relationships, can be viewed as a complex network with concepts as …

Crowdsourcing Atop Blockchains, Yuan Lu

Dissertations

Traditional crowdsourcing systems, such as Amazon's Mechanical Turk (MTurk), though once acquiring great economic successes, have to fully rely on third-party platforms to serve between the requesters and the workers for basic utilities. These third-parties have to be fully trusted to assist payments, resolve disputes, protect data privacy, manage user authentications, maintain service online, etc. Nevertheless, tremendous real-world incidents indicate how elusive it is to completely trust these platforms in reality, and the reduction of such over-reliance becomes desirable.

In contrast to the arguably vulnerable centralized approaches, a public blockchain is a distributed and transparent global "consensus computer" that is …

Mathematical Models And Tools To Understand Coupled Circadian Oscillations And Limit Cycling Systems, Guangyuan Liao

Dissertations

The circadian rhythm refers to an internal body process that regulates many body processes including the sleep-wake cycle, digestion and hormone release. The ability of a circadian system to entrain to the 24-hour light-dark cycle is one of the most important properties. There are several scenarios in which circadian oscillators do not directly receive light-dark forcing. Instead they are part of hierarchical systems in which, as \peripheral" oscillators, they are periodically forced by other \central" circadian oscillators that do directly receive light input. Such dynamics are modeled as hierarchical coupled limit cycle systems. Those models usually have a large population, …

Resonant Triad Interactions In One And Two-Layer Systems, Malik Chabane

Dissertations

This dissertation is a study of the weakly nonlinear resonant interactions of a triad of gravity-capillary waves in systems of one and two fluid layers of arbitrary depth, in one and two-dimentions. For one-layer systems, resonant triad interactions of gravity-capillary waves are considered and a region where resonant triads can be always found is identified, in the two-dimensional wavevector angles-space. Then a description of the variations of resonant wavenumbers and wave frequencies over the resonance region is given. The amplitude equations correct to second order in wave slope are used to investigate special resonant triads that, providing their initial amplitude …

Machine Learning For Scientific Data Mining And Solar Eruption Prediction, Hao Liu

Dissertations

This dissertation explores new machine learning techniques and adapts them to mine scientific data, specifically data from solar physics and space weather studies. The dissertation tackles three important problems in heliophysics: solar flare prediction, coronal mass ejection (CME) prediction and Stokes inversion.

First, the dissertation presents a long short-term memory (LSTM) network for predicting whether an active region (AR) would produce a certain class of solar flare within the next 24 hours. The essence of this approach is to model data samples in an AR as time series and use LSTMs to capture temporal information of the data samples. The …

Efficient Approximations For Stationary Single-Channel Calcium Nanodomains, Yinbo Chen

Dissertations

Mathematical and computational modeling plays an important role in the study of local Ca2+ signals underlying many fundamental physiological processes such as synaptic neurotransmitter release and myocyte contraction. Closed-form approximations describing steady-state distribution of Ca2+ in the vicinity of an open Ca2+ channel have proved particularly useful for the qualitative modeling of local Ca2+ signals. This dissertation presents several simple and efficient approximants for the equilibrium Ca2+ concentration near a point source in the presence of a mobile Ca2+ buffer, which achieve great accuracy over a wide range of model parameters. Such approximations provide an efficient method for estimating Ca2+ …

Energy And Performance-Optimized Scheduling Of Tasks In Distributed Cloud And Edge Computing Systems, Haitao Yuan

Dissertations

Infrastructure resources in distributed cloud data centers (CDCs) are shared by heterogeneous applications in a high-performance and cost-effective way. Edge computing has emerged as a new paradigm to provide access to computing capacities in end devices. Yet it suffers from such problems as load imbalance, long scheduling time, and limited power of its edge nodes. Therefore, intelligent task scheduling in CDCs and edge nodes is critically important to construct energy-efficient cloud and edge computing systems. Current approaches cannot smartly minimize the total cost of CDCs, maximize their profit and improve quality of service (QoS) of tasks because of aperiodic arrival …

Flare Emission Observed In High Resolution And Comparison With Numerical Modeling, Nengyi Huang

Dissertations

As one of the most intense activities on the solar surface, flares have been extensively observed and studied ever since the first report. The standard model of solar flares has been established and commonly accepted. However, many limitations from the researching tools have left some of the problems unsolved or controversial. For example, the density of electrons in the corona is lower than it is required to activate the observed emission in HXR, and the mechanism that these electron beams can penetrate down to lower chromosphere is unclear. Many theoretical scenarios were suggested, and more observations had been in need. …

Global Optimization Algorithms For Image Registration And Clustering, Cuicui Zheng

Dissertations

Global optimization is a classical problem of finding the minimum or maximum value of an objective function. It has applications in many areas, such as biological image analysis, chemistry, mechanical engineering, financial analysis, deep learning and image processing. For practical applications, it is important to understand the efficiency of global optimization algorithms. This dissertation develops and analyzes some new global optimization algorithms and applies them to practical problems, mainly for image registration and data clustering.

First, the dissertation presents a new global optimization algorithm which approximates the optimum using only function values. The basic idea is to use the points …

Changing The Focus: Worker-Centric Optimization In Human-In-The-Loop Computations, Mohammadreza Esfandiari

Dissertations

A myriad of emerging applications from simple to complex ones involve human cognizance in the computation loop. Using the wisdom of human workers, researchers have solved a variety of problems, termed as “micro-tasks” such as, captcha recognition, sentiment analysis, image categorization, query processing, as well as “complex tasks” that are often collaborative, such as, classifying craters on planetary surfaces, discovering new galaxies (Galaxyzoo), performing text translation. The current view of “humans-in-the-loop” tends to see humans as machines, robots, or low-level agents used or exploited in the service of broader computation goals. This dissertation is developed to shift the focus back …

An Automated Feedback System To Support Student Learning Of Conceptual Knowledge In Writing-To-Learn Activities, Ye Xiong

Dissertations

As a pedagogical strategy, Writing-to-Learn (WTL) intends to use writing to improve students’ understanding of course content. However, most of the existing feedback systems for writing are mainly focused on improving students’ writing skills rather than their conceptual development. In this dissertation, an automatic approach is proposed to generate timely, actionable, and individualized feedback based on comparing knowledge representations extracted from lecture slides and individual students’ writing assignments. The novelty of the proposed approach lies in the feedback generation: to help students assimilate new knowledge into their existing knowledge better, their current knowledge is modeled as a set of matching …

1,4-Dioxane Biodegradation In Propanotrophs: Molecular Foundations And Implications For Environmental Remediation, Li Fei

Dissertations

1,4-Dioxane (dioxane) has emerged with an escalating concern given its human carcinogenicity and widespread occurrence in groundwater. Bioremediation is promising as an effective and cost-efficient treatment alternative for in situ or ex situ cleanup of dioxane and co-existing pollutants in the field. Soluble di-iron monooxygenases (SDIMOs) are reputed for their essential roles in initiating the cleavage of dioxane and other pollutants. In this doctoral dissertation, molecular foundations for SDIMOs-mediated dioxane biodegradation are untangled to promote the development and implication of site-specific bioremediation and natural attenuation strategies. This dissertation focused on propanotrophic bacteria given their pivotal roles in dioxane metabolism and …

The Effect Of Phytoplankton Properties On The Ingestion Of Marine Snow By Calanus Pacificus, Grace Cawley

Theses

The aggregation of phytoplankton into marine snow provides a mechanism by which smaller particles can coagulate to form larger particles, which can be consumed at various depths or readily transported to the deep ocean and sequestered from the atmosphere on time scales of a thousand years or more. Zooplankton interacting with these large carbon-rich aggregates can obtain nutrition in environments where the phytoplankton size spectrum is small and not directly available, enhancing the possibility of obtaining adequate nutrition in environments dominated by small cells. In addition, interactions between zooplankton and marine snow can result in fragmentation, thus affecting the particle …