Physical Sciences and Mathematics Commons^™

Resource Aware Clustering For Tackling The Heterogeneity Of Participants In Federated Learning, Rahul Mishra, Hari Prabhat Gupta, Garvit Banga, Sajal K. Das

Computer Science Faculty Research & Creative Works

Federated Learning Is A Training Framework That Enables Multiple Participants To Collaboratively Train A Shared Model While Preserving Data Privacy. The Heterogeneity Of Devices And Networking Resources Of The Participants Delay The Training And Aggregation. The Paper Introduces A Novel Approach To Federated Learning By Incorporating Resource-Aware Clustering. This Method Addresses The Challenges Posed By The Diverse Devices And Networking Resources Among Participants. Unlike Static Clustering Approaches, This Paper Proposes A Dynamic Method To Determine The Optimal Number Of Clusters Using Dunn Indices. It Enables Adaptability To The Varying Heterogeneity Levels Among Participants, Ensuring A Responsive And Customized Approach To …

Robust Bidirectional Poly-Matching, Ween Jiann Lee, Maksim Tkachenko, Hady Wirawan Lauw

Research Collection School Of Computing and Information Systems

A fundamental problem in many scenarios is to match entities across two data sources. It is frequently presumed in prior work that entities to be matched are of comparable granularity. In this work, we address one-to-many or poly-matching in the scenario where entities have varying granularity. A distinctive feature of our problem is its bidirectional nature, where the 'one' or the 'many' could come from either source arbitrarily. Moreover, to deal with diverse entity representations that give rise to noisy similarity values, we incorporate novel notions of receptivity and reclusivity into a robust matching objective. As the optimal solution to …

Application Of The Two-Variable Model To Simulate A Multisensory Reaction-Time Task, Rebecca Brady, John Butler

Academic Posters Collection

To navigate the world in an efficient manner, the brain seamlessly integrates signals received across multiple sensory modalities. Behavioral studies have suggested that multisensory processing is a winner-take-all sensory response mechanism to some optimal combination of sensory signals. In addition, multiple sensory cues are not always beneficial with some studies showing maladaptive multisensory processing as an identifier of older adults prone to falls from age matched healthy controls.

A stalwart of modelling sensory decision-making is the work by (Wong &Wang, 2006) but to date almost all of this research has been focused on unisensory tasks. We extend the reduced two-variable …

Active Learning Strategy For Covid-19 Annotated Dataset, Amril Nazir, Ricky Maulana Fajri

All Works

The efficient diagnosis of COVID-19 plays a key role in preventing its spread. Recently, many artificial intelligence techniques, such as the deep neural network approach, have been implemented to help efficient diagnosis of COVID-19. However, the accurate performance of deep learning depends on the tuning of many hyperparameters and a large amount of labeled data. This COVID-19 data bottleneck also leads to insufficient human resources for data labeling, which presents a challenging obstacle. In this paper, a novel discriminative batch-mode active learning (DS3) is proposed to allow faster and more effective COVID-19 data annotation. The framework specifically designed to suit …

Selectivity And Robustness Of Sparse Coding Networks, Dylan M. Paiton, Charles Frye, Sheng Y. Lundquist, Joel D. Bowen, Ryan Zarcone, Bruno A. Olshausen

Computer Science Faculty Publications and Presentations

We investigate how the population nonlinearities resulting from lateral inhibition and thresholding in sparse coding networks influence neural response selectivity and robustness. We show that when compared to pointwise nonlinear models, such population nonlinearities improve the selectivity to a preferred stimulus and protect against adversarial perturbations of the input. These findings are predicted from the geometry of the single-neuron iso-response surface, which provides new insight into the relationship between selectivity and adversarial robustness. Inhibitory lateral connections curve the iso-response surface outward in the direction of selectivity. Since adversarial perturbations are orthogonal to the iso-response surface, adversarial attacks tend to be …

Models As Weapons: Review Of Weapons Of Math Destruction: How Big Data Increases Inequality And Threatens Democracy By Cathy O’Neil (2016), Samuel L. Tunstall

Numeracy

Cathy O’Neil. 2016. Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy (New York, NY: Crown) 272 pp. ISBN 978-0553418811.

Accessible to a wide readership, Cathy O’Neil’s Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy provides a lucid yet alarming account of the extensive reach of mathematical models in influencing all of our lives. With a particular eye towards social justice, O’Neil not only warns modelers to be cognizant of the effects of their work on real people—especially vulnerable groups who have less power to fight back—but also encourages laypersons to take initiative …

Differential Equations Models Of Pathogen-Induced Single- And Multi-Organ Tissue Damage, Fiona Lynch

Honors Theses

The rise of antibiotic resistance has created a significant burden on healthcare systems around the world. Antibiotic resistance arises from the increased use of antibiotic drugs and antimicrobial agents, which kill susceptible bacterial strains, but have little effect on strains that have a mutation allowing them to survive antibiotic treatment, defined as “resistant” strains. With no non-resistant bacteria to compete for resources, the resistant bacteria thrives in this environment, continuing to reproduce and infect the host with an infection that does not respond to traditional antibiotic treatment.

A number of strategies have been proposed to tackle the problem of antibiotic …

A Multi-Objective Approach To Tactical Maneuvering Within Real Time Strategy Games, Christopher D. Ball

Theses and Dissertations

The real time strategy (RTS) environment is a strong platform for simulating complex tactical problems. The overall research goal is to develop artificial intelligence (AI) RTS planning agents for military critical decision making education. These agents should have the ability to perform at an expert level as well as to assess a players critical decision-making ability or skill-level. The nature of the time sensitivity within the RTS environment creates very complex situations. Each situation must be analyzed and orders must be given to each tactical unit before the scenario on the battlefield changes and makes the decisions no longer relevant. …

Big Data: Immediate Opportunities And Longer Term Challenges, Jens Pohl, Kym Jason Pohl

Collaborative Agent Design (CAD) Research Center

The transformation of words, locations, and human interactions into digital data forms the basis of trend detection and information extraction opportunities that can be automated with the increasing availability of relatively inexpensive computer storage and processing technology. Trend detection, which focuses on what, is facilitated by the ability to apply analytics to an entire corpus of data instead of a random sample. Since the corpus essentially includes all data within a population there is no need to apply any of the precautions that are in order to ensure the representativeness of a sample in traditional statistical analysis. Several examples are …

Water Resources Systems Planning And Management, Piro Zoga, Aida Bode, Ramiz Balla, Nevton Kodhelaj, Spartak Kucaj

UBT International Conference

The most fundamental human needs of water are for drinking, cooking and personal hygiene. In order to fulfill these needs, the quality of the water used should cause no risk to human health. The quality of the water in nature also affects the condition of ecosystems all living organisms depend on. At the same time, humans use water bodies as convenient sinks for the disposal of domestic, industrial and agricultural wastewaters. This, of course, degrades the quality of those water bodies. Water resources management involves the monitoring and management of water quality as much as the monitoring and management of …

Modeling Ice Streams, Aitbala Sargent

Electronic Theses and Dissertations

Modeling glacier and ice sheet flow is a computationally challenging problem. The most challenging part in simulating ice sheet flow is modeling the fastest moving part of ice sheets, ice streams. In the first part of the thesis, we have constructed two numerical models of isothermal ice stream flow, a three-dimensional full-Stokes ice-sheet/ice-stream/ice-shelf model and a modified MacAyeal-Morland ice-stream/ice-shelf model. In the second part of the thesis, we studied the possibility of using SuperLU-DIST multiprocessor software package for solving the systems of linear equations generated by the model. The uniqueness of the modified MacAyeal-Morland model is in its inclusion of …

A Mathematical Model For Simplifying Representations Of Objects In A Geographic Information System, Gabriel Perrow

Electronic Theses and Dissertations

The study of operations on representations of objects is well documented in the realm of spatial engineering. However, the mathematical structure and formal proof of these operational phenomena are not thoroughly explored. Other works have often focused on query-based models that seek to order classes and instances of objects in the form of semantic hierarchies or graphs. In some models, nodes of graphs represent objects and are connected by edges that represent different types of coarsening operators. This work, however, studies how the coarsening operator "simplification" can manipulate partitions of finite sets, independent from objects and their attributes. Partitions that …

Visualization And Animation Of A Missile/Target Encounter, Jeffrey T. Bush

Theses and Dissertations

Existing missile/target encounter modeling and simulation systems focus on improving probability of kill models. Little research has been done to visualize these encounters. These systems can be made more useful to the engineers by incorporating current computer graphics technology for visualizing and animating the encounter. Our research has been to develop a graphical simulation package for visualizing both endgame and full fly-out encounters. Endgame visualization includes showing the interaction of a missile, its fuze cone proximity sensors, and its target during the final fraction of a second of the missile/target encounter. Additionally, this system displays dynamic effects such as the …

Images 2.1 : An Integrated Model Of An Arid Grazing System, Z G. Yan, K M. Wang

Resource management technical reports

The name IMAGES stands for An Integrated Model of an Arid Grazing Ecological System. The model was initiated by Dr R. Hacker in 1987-8 as an activity to (1) evaluate alternative management strategies and (2) identify key ecological processes and research priorities in shrub rangelands of Western Australia. Version 1 of the model was published in Agriculture Systems in 1991 (Hacker et al. 1991) and here after will be referred to as IMAGES 1. IMAGES 1 is a vegetation model, capable of predicting the prob ility of recruitment and mortality of the desirable species in a given vegetation type under …

Visibility-Related Problems On Parallel Computational Models, Himabindu Gurla

Computer Science Theses & Dissertations

Visibility-related problems find applications in seemingly unrelated and diverse fields such as computer graphics, scene analysis, robotics and VLSI design. While there are common threads running through these problems, most existing solutions do not exploit these commonalities. With this in mind, this thesis identifies these common threads and provides a unified approach to solve these problems and develops solutions that can be viewed as template algorithms for an abstract computational model. A template algorithm provides an architecture independent solution for a problem, from which solutions can be generated for diverse computational models. In particular, the template algorithms presented in this …

Multiple Learner Systems Using Resampling Methods, Binyun Xie

Computer Science Theses & Dissertations

The N-Learners Problem deals with combining a number of learners such that the resultant system is "better", under some criterion, than the best of the individual learners. We consider a system of probably approximately correct concept learners. Depending on the available information, there are several methods to make the composite system better than the best of the individual learners. If a sample and an oracle that generates data points (but, not their classification) is available, then we show that we can achieve arbitrary levels of the normalized confidence of the composite system if (a) a robust learning algorithm is available, …

Study Of The Effectiveness Of Cost-Estimation Models And Complexity Metrics On Small Projects, Lein-Lein Chen

FIU Electronic Theses and Dissertations

Software cost overruns and time delay are common occurrences in the software development process. To reduce the occurrences of these problems, software cost estimation models and software complexity metrics measurements are two popular approaches used by the industry.

Most of the related studies are conducted for large scale software projects. In this thesis, we have investigated the effectiveness of three popular cost estimation models and program complexity metrics in so far as their applicability to small scale projects is concerned.

Experiments conducted on the programs collected from FIU and NCR corporation indicate that none of the cost estimation models precisely …

Decide : How Much Superphosphate, G A. Robertson, J. W. Bowden, N. J. Halse

Journal of the Department of Agriculture, Western Australia, Series 4

* A 400 per cent increase in ihe price of superphosphate has reduced the economic optimum rates of super for crops and pastures in 1975.

* Many factors, both biological and economic, must be taken into account in determining the rate of superphosphate to apply.

• DECIDE, a model developed by CSIRO and the Department of Agriculture, provides a formal system in which all these factors can be considered.

• DECIDE is based on the results of all research on superphosphate carried out in Western Australia. However, each farmer's own knowledge of his farm, the soils, crops and animals is …