Physical Sciences and Mathematics Commons^™

Characterization And Automatic Updates Of Deprecated Machine-Learning Api Usages, Stefanus Agus Haryono, Thung Ferdian, David Lo, Julia Lawall, Lingxiao Jiang

Research Collection School Of Computing and Information Systems

Due to the rise of AI applications, machine learning (ML) libraries, often written in Python, have become far more accessible. ML libraries tend to be updated periodically, which may deprecate existing APIs, making it necessary for application developers to update their usages. In this paper, we build a tool to automate deprecated API usage updates. We first present an empirical study to better understand how updates of deprecated ML API usages in Python can be done. The study involves a dataset of 112 deprecated APIs from Scikit-Learn, TensorFlow, and PyTorch. Guided by the findings of our empirical study, we propose …

Biasheal: On-The-Fly Black-Box Healing Of Bias In Sentiment Analysis Systems, Zhou Yang, Harshit Jain, Jieke Shi, Muhammad Hilmi Asyrofi, David Lo

Research Collection School Of Computing and Information Systems

Although Sentiment Analysis (SA) is widely applied in many domains, existing research has revealed that the unfairness in SA systems can be harmful to the welfare of less privileged people. Several works propose pre-processing and in-processing methods to eliminate bias in SA systems, but little attention is paid to utilizing post-processing methods to heal bias. Postprocessing methods are particularly important for systems that use third-party SA services. Systems that use such services have no access to the SA engine or its training data and thus cannot apply pre-processing nor in-processing methods. Therefore, this paper proposes a black-box post-processing method to …

Secure And Verifiable Outsourced Data Dimension Reduction On Dynamic Data, Zhenzhu Chen, Anmin Fu, Robert H. Deng, Ximeng Liu, Yang Yang, Yinghui Zhang

Research Collection School Of Computing and Information Systems

Dimensionality reduction aims at reducing redundant information in big data and hence making data analysis more efficient. Resource-constrained enterprises or individuals often outsource this time-consuming job to the cloud for saving storage and computing resources. However, due to inadequate supervision, the privacy and security of outsourced data have been a serious concern to data owners. In this paper, we propose a privacypreserving and verifiable outsourcing scheme for data dimension reduction, based on incremental Non-negative Matrix Factorization (NMF) method. We emphasize the importance of incremental data processing, exploiting the properties of NMF to enable data dynamics in consideration of data updating …

The Empathetic Car: Exploring Emotion Inference Via Driver Behaviour And Traffic Context, Shu Liu, Kevin Koch, Zimu Zhou, Simon Foll, Xiaoxi He, Tina Menke, Elgar Fleisch, Felix Wortmann

Research Collection School Of Computing and Information Systems

An empathetic car that is capable of reading the driver’s emotions has been envisioned by many car manufacturers. Emotion inference enables in-vehicle applications to improve driver comfort, well-being, and safety. Available emotion inference approaches use physiological, facial, and speech-related data to infer emotions during driving trips. However, existing solutions have two major limitations: Relying on sensors that are not built into the vehicle restricts emotion inference to those people leveraging corresponding devices (e.g., smartwatches). Relying on modalities such as facial expressions and speech raises privacy concerns. By contrast, researchers in mobile health have been able to infer affective states (e.g., …

Galaxy And Mass Assembly (Gama), U. Sureshkumar, A. Durkalec, A. Pollo, M. Bilicki, J. Loveday, D. J. Farrow, Benne Holwerda, A. M. Hopkins, J. Liske, K. A. Pimbblet, E. N. Taylor, A. H. Wright

Faculty and Staff Scholarship

Context. Galaxies are biased tracers of the underlying network of dark matter. The strength of this bias depends on various galaxy properties and on redshift. One of the methods used to study these dependences of the bias is measurement of galaxy clustering. Such studies are made using galaxy samples from various catalogues, which frequently bear their own problems related to sample selection methods. It is therefore crucial to understand how sample choice influences clustering measurements and which galaxy property is the most direct tracer of the galaxy environment.

Aims. We investigate how different galaxy properties, such as luminosities in the …

The Impact Of Laser Control On The Porosity And Microstructure Of Selective Laser Melted Nickel Superalloy 718, Travis E. Shelton, Gregory R. Cobb, Carl R. Hartsfield, Benjamin M. Doane, Cayla C. Eckley, Ryan A. Kemnitz

Faculty Publications

Additively manufacturing high performance metals by laser processing represents an exciting opportunity to exploit localized properties by varying input parameters throughout the process. This work explores the solidification and microstructural properties of selectively laser melted (SLM) Inconel 718 (IN718) using unique processing parameters. By employing traditional pulsed laser physics techniques, samples were manufactured with a continuous wave laser to study a potential ubiquitous approach. While the overall power density was controlled, the power, speed, and hatch spacing were varied. The porosity and grain sizes of the samples were characterized by optical and scanning electron microscopes. The influence of processing parameters …

Clifford Harmonics, Samuel L. Hosmer

Dissertations, Theses, and Capstone Projects

In 1980 Michelsohn defined a differential operator on sections of the complex Clifford bundle over a compact Kähler manifold M. This operator is a differential and its Laplacian agrees with the Laplacian of the Dolbeault operator on forms through a natural identification of differential forms with sections of the Clifford bundle. Relaxing the condition that M be Kähler, we introduce two differential operators on sections of the complex Clifford bundle over a compact almost Hermitian manifold which naturally generalize the one introduced by Michelsohn. We show surprising Kähler- like symmetries of the kernel of the Laplacians of these operators in …

Novel Hybrid Resampling Algorithms For Parallel/Distributed Particle Filters, Xudong Zhang

Dissertations, Theses, and Capstone Projects

Particle filters, also known as sequential Monte Carlo (SMC) methods, use the Bayesian inference and the stochastic sampling technique to estimate the states of dynamic systems from given observations. Parallel/Distributed particle filters were introduced to improve the performance of sequential particle filters by using multiple processing units (PUs). The classical resampling algorithm used in parallel/distributed particle filters is a centralized scheme, called centralized resampling, which needs a central unit (CU) to serve as a hub for data transfers. As a result, the centralized resampling procedures produce extra communication costs, which lowers the speedup factors in parallel computing. Even though some …

A Geometric Model For Real And Complex Differential K-Theory, Matthew T. Cushman

Dissertations, Theses, and Capstone Projects

We construct a differential-geometric model for real and complex differential K-theory based on a smooth manifold model for the K-theory spectra defined by Behrens using spaces of Clifford module extensions. After writing representative differential forms for the universal Pontryagin and Chern characters we transgress these forms to all the spaces of the spectra and use them to define an abelian group structure on maps up to an equivalence relation that refines homotopy. Finally we define the differential K-theory functors and verify the axioms of Bunke-Schick for a differential cohomology theory.

Tools And Strategies For The Patterning Of Bioactive Molecules And Macromolecules, Daniel J. Valles

Dissertations, Theses, and Capstone Projects

Hypersurface Photolithography (HP) is a printing method for fabricating structures and patterns composed of soft materials bound to solid surfaces and with ~1 micrometer resolution in the x, y, and z dimensions. This platform leverages benign, low intensity light to perform photochemical surface reactions with spatial and temporal control of irradiation, and, as a result, is particularly useful for patterning delicate organic and biological material. In particular, surface- initiated controlled radical polymerizations can be leveraged to create arbitrary polymer and block- copolymer brush patterns. Chapter 1 will review the advances in instrumentation architectures from our group that have made these …

Nonlinear Optical Studies Of Interfacial Ferroelectricity And Strain Distribution In Perovskite Dielectric Films, Tony Le

Dissertations, Theses, and Capstone Projects

Dielectric and ferroelectric perovskite films have been model energy storage structures for their low-dielectric loss, extremely high charge-discharge speed, and good temperature stability, yet there is still much to understand about the material’s limitations. This dissertation presents a detailed understanding of the strain-induced ferroelectricity at the boundary between a strontium titanate (SrTiO₃) ultrathin film epitaxially grown on a germanium (Ge) substrate through optical second harmonic generation (SHG), and the polydomain distribution in the Zr-doped BaTiO₃ (BZT) films by time-resolved pump-probe spectroscopy.

First, SHG measurements were performed to reveal interfacial ferroelectricity in the epitaxial SrTiO₃/Ge (100) …

Logics Of Resource And Justification, Hirohiko Kushida

Dissertations, Theses, and Capstone Projects

It is a well-known result by G ̈odel in 1933 that the Intuitionistic Logic can be embedded into a system which is essentially equivalent to the modal logic S4. This can be considered to be an attempt to provide a provability semantics to the Intuitionistic Logic. This work had caused some problems: the exact arithmetical meaning of the Intuitionistic Logic and S4, the exact axiomatization of formal provability in formal arithmetic and the standard model of arithmetic. These days the arithmetical interpretation has been extended and generalized to epistemological interpretation for various modal logics, which resulted in various systems of …

Topological Classical Wave Systems With Modulations, Interactions, And Higher-Order Topological States, Mengyao Li

Dissertations, Theses, and Capstone Projects

Topological phases in classical wave systems, such as photonic and acoustic, have been actively investigated and applied for wave guiding, lasing, and numerous other novel phenomena and device applications Topological phase transitions enable robust boundary states, and the field has been broadening recently into a vast variety of systems with temporal modulations and interactions. Floquet modulation, for example, is the modulation applied periodically in time which may break symmetries and leads to novel topological phases.

Introducing non-Hermitian Floquet modulation enables more interesting phenomena including bandgap in imaginary part of the spectrum and gainy/lossy topological edge states with complex energy values. …

Piecewise Linear Manifold Clustering, Artyom Diky

Dissertations, Theses, and Capstone Projects

This work studies the application of topological analysis to non-linear manifold clustering. A novel method, that exploits the data clustering structure, allows to generate a topological representation of the point dataset. An analysis of topological construction under different simulated conditions is performed to explore the capabilities and limitations of the method, and demonstrated statistically significant improvements in performance. Furthermore, we introduce a new information-theoretical validation measure for clustering, that exploits geometrical properties of clusters to estimate clustering compressibility, for evaluation of the clustering goodness-of-fit without any prior information about true class assignments. We show how the new validation measure, when …

Developing Dynamical Probes Of Quantum Spin Liquids Inspired By Techniques From Spintronics, Joshua Aftergood

Dissertations, Theses, and Capstone Projects

We theoretically study low dimensional insulating spin systems using spin fluctuations as a probe of the spin dynamics. In some systems, low dimensionality in conjunction with other quantum fluctuation enhancing effects impedes spontaneous generation of long range magnetic ordering down to zero absolute temperature. In particular, we focus on exotic spin systems hosting mobile, fractionalized spin excitations above their ground state, and ultimately show that techniques already commonplace in spintronics can be utilized in the context of quantum magnetism to develop probes of exotic ground states.

We initially consider quantum spin chains (QSCs), and examine a system of two exchange-coupled …

Molecular Dynamics Simulations Of Self-Assemblies In Nature And Nanotechnology, Phu Khanh Tang

Dissertations, Theses, and Capstone Projects

Nature usually divides complex systems into smaller building blocks specializing in a few tasks since one entity cannot achieve everything. Therefore, self-assembly is a robust tool exploited by Nature to build hierarchical systems that accomplish unique functions. The cell membrane distinguishes itself as an example of Nature’s self-assembly, defining and protecting the cell. By mimicking Nature’s designs using synthetically designed self-assemblies, researchers with advanced nanotechnological comprehension can manipulate these synthetic self-assemblies to improve many aspects of modern medicine and materials science. Understanding the competing underlying molecular interactions in self-assembly is always of interest to the academic scientific community and industry. …

Monte Carlo, Molecular Dynamics And Network Analysis Of The Gramicidin Water Channel And Proton Transfer Pathways To Qb In Photosynthetic Reaction Centers, Yingying Zhang

Dissertations, Theses, and Capstone Projects

Water molecules play a key role in all biochemical processes. They help define the shape of proteins, and they are reactant or product in many reactions and are released as ligands are bound. They facilitate transfer of protons through transmembrane proton channel, pump and transporter proteins. Continuum electrostatics (CE) force fields such as used in MCCE (Multi-Conformation Continuum Electrostatics) capture electrostatic interactions in biomolecules with an implicit solvent, to give the averaged solvent water equilibrium properties. Hybrid CE methods can use explicit water molecules within the protein surrounded by implicit solvent. These hybrid methods permit the study of explicit hydrogen …

A Unified Approach To M Dwarf Ages, Rocio Kiman

Dissertations, Theses, and Capstone Projects

Estimating ages of M dwarfs, the lowest mass stars in the Galaxy, is a current unresolved problem in Astrophysics. This dissertation focuses on developing a method to estimate ages for M dwarfs. This method consists of combining different age indicators in a Bayesian framework to achieve a precise age measurement. The age indicators I use are: 1) position in the color-magnitude diagram which is a proxy for effective temperature and luminosity, 2) movement of the stars in the Galaxy, or its 3D kinematics, 3) magnetic activity, measured by the Halpha emission line of the spectrum. In this dissertation I analyze …

Carbon And Nitrogen Dynamics Of A Lagoon Salt Marsh In Davenport, California, Mark D. Gormley

Dissertations, Theses, and Capstone Projects

Wetlands are complex environments that play a critical role in regulating the global biogeochemical cycle of carbon (C) and nitrogen (N). Wetlands are critical contributors to global climate change and atmospheric chemistry since they store as much as 33% of the world’s soil organic carbon (SOC), release more than 20% of the atmospheric methane (CH₄), and produce nitrous oxide (N₂O), an extremely potent greenhouse gas (GHG). Despite the enormous radiative forcing potential of carbon dioxide (CO₂), CH₄ and N₂O derived from wetlands, uncertainties over the rates of C sequestration and GHG …

Solving Multiple Inference In Graphical Models, Cong Chen

Dissertations, Theses, and Capstone Projects

For inference problems in graphical models, much effort has been directed at algorithms for obtaining one single optimal prediction. In practice, the data is often noisy or incomplete, which makes one single optimal solution unreliable. To address this problem, multiple Inference is proposed to find several best solutions, M-Best, where multiple hypotheses are preferred for advanced reasoning. People use oracle accuracy as an evaluation criterion expecting one of the solutions has high accuracy with the ground truth. It has been shown that it is beneficial for the top solutions to be diverse. Approaches for solving diverse multiple inference are proposed …

Evaluation Of The Potential Of Automated Sem-Eds Analysis For The Discrimination Of Inorganic Soil Particles, Anna S. Duggar

Dissertations, Theses, and Capstone Projects

Soil, that complex mixture of minerals, organic particles, chemicals, and anthropogenic materials, is ubiquitous and easily transferred, which gives it forensic relevance. Forensic soil analysis has traditionally included the identification and comparison of both organic and inorganic components, the inorganic portion being most commonly examined by light microscopy and other instrumental techniques. Predominantly, the inorganic materials in soil are naturally-occurring minerals, contributed as grains from the surrounding bedrock, which gives a soil its regional character. That parent material then undergoes physical, chemical, and anthropogenic changes producing wide geographic variation within a single region. This variation forms the basis for potential …

Representing The Derivative Of Trace Of Holonomy, Jeffrey Peter Kroll

Dissertations, Theses, and Capstone Projects

Trace of holonomy around a fixed loop defines a function on the space of unitary connections on a hermitian vector bundle over a Riemannian manifold. Using the derivative of trace of holonomy, the loop, and a flat unitary connection, a functional is defined on the vector space of twisted degree 1 cohomology classes with coefficients in skew-hermitian bundle endomorphisms. It is shown that this functional is obtained by pairing elements of cohomology with a degree 1 homology class built directly from the loop and equipped with a flat section obtained from the variation of holonomy around the loop. When the …

Don't Sell Them Short, There's More To Bacterial Natural Products Than Antibiotics, Alison Clare Domzalski

Dissertations, Theses, and Capstone Projects

Recent genomic studies of microbiomes have revealed an overwhelming number of biosynthetic genes of unknown function. Most of these “cryptic” biosynthetic genes are not expressed in laboratory monocultures of individual microbes. Thus, there remains tremendous untapped potential for natural products discovery. Here we employ mixed microbial culture (MMC) as a simple yet powerful approach to awaken cryptic biosynthetic gene clusters. Our preliminary studies demonstrated that arrays of metabolites could be induced in MMCs upon environmental cues, such as surface adhesion. Using this system, we have screened, identified, and isolated bioactive bacterial metabolites, which were characterized structurally and biologically. Of the …

Connecting Communities To Coastal Resilience: Enhancing Sustainability Through Public Participation In Salt Marsh Management And Restoration In Suffolk County, Ny, Jennifer L. Mcgivern

Dissertations, Theses, and Capstone Projects

Coastal resiliency is becoming significantly more critical to the livelihood of coastal communities as the frequency and intensity of storm events increases and is exacerbated by rising sea levels due to climate change. In October 2012 Superstorm Sandy impacted the New York-New Jersey area costing over $70 billion in storm damages and 147 lives lost, as storm surges surpassed record highs for the region. Protruding more than 100 miles into the Atlantic Ocean with over 1,000 miles of shoreline, Long Island is particularly vulnerable to the increasingly ferocious and numerous storms as well as the rising sea levels that climate …

Wave Excitation And Dynamics In Disordered Systems, Yiming Huang

Dissertations, Theses, and Capstone Projects

This thesis presents studies of the field and energy excited in disordered systems as well as the dynamics of scattering.

Dynamic and steady state aspects of wave propagation are deeply connected in lossless open systems in which the scattering matrix is unitary. There is then an equivalence among the energy excited within the medium through all channels, the Wigner time delay, which is the sum of dwell times in all channels coupled to the medium, and the density of states. But these equivalences fall away in the presence of material loss or gain. In this paper, we use microwave measurements, …

Linear And Non Linear Properties Of Two-Dimensional Exciton-Polaritons, Mandeep Khatoniar

Dissertations, Theses, and Capstone Projects

Technology has been accelerating at breakneck speed since the first quantum revolution, an era that ushered transistors and lasers in the late 1940s and early 1960s. Both of these technologies relied on a matured understanding of quantum theories and since their inception has propelled innovation and development in various sectors like communications, metrology, and sensing. Optical technologies were thought to be the game changers in terms of logic and computing operations, with the elevator pitch being "computing at speed of light", a fundamental speed limit imposed by this universe’s legal system (a.k.a physics). However, it was soon realized that that …

Synthesis, Characterization And Applications Of Europium Based Perovskite Oxide Nanoparticles, Nasim Farahmand

Dissertations, Theses, and Capstone Projects

The synthesis and characterization of nanocrystalline transition and rare-earth metal oxides have been studied widely over the past years. These materials possess unique physical, chemical, magnetic and dielectric properties which make them potential candidates for a wide range of applications such as electronics, energy storage and photocatalysis. There are a number of synthesis methods for preparing these materials, where most of them require stringent conditions such as high pressure and high temperature. However, in this study, a modified sol-gel synthesis has been used with a lower calcination temperature to make uniform nanocrystals of the transition/rare-earth metal oxide family. The focus …

Adversarial Training For Skill Learning In A Mobile Robot, Todd W. Flyr

Dissertations, Theses, and Capstone Projects

Machine Learning in mobile robotics is sometimes hampered by the difficulties associated with the creation of a large corpus of labeled data that most neural network based learning algorithms demand. In recent years, advances in the field of machine learning have been facilitated via the creation of large collaboratively-created labeled training datasets that researchers can use as the basis for experiments to validate and improve their candidate neural network architectures. For the field of robotics, however, tasks are so disparate and the physical devices so varied that in most cases the creation of collaborative benchmark datasets are impractical. Obtaining data …

Development Of An Lcms Analytical Method With Applicability To Determination Of Trace Level Explosives In Environmental Waters, Cristina I. Veresmortean

Dissertations, Theses, and Capstone Projects

Traces of explosives and their degradation products present in explosives contaminated sites from military sources, such as training facilities, impose a great concern over the public health and our eco-system. This work is centered around finding an efficient way to extraction of these chemicals from water samples and their analysis using advanced separation technology: LC-MS/MS Shimadzu Prominence 8050 and AB Sciex Trap 4000 LC-MS/MS. Different extraction methods will be proposed, along with chromatographic and mass spectroscopy method development. Dependent on the target analyte polarity, thermal degradation, solubility and type of matrix, an appropriate method of analysis will be established. The …

Automated Code Engine For Tensor Hypercontraction: Derivation, Optimization And Implementation Of Rank-Reduced Coupled Cluster Theories, Yao Zhao

Dissertations, Theses, and Capstone Projects

The ultimate goal of electronic structure theory is solving the electronic Schr¨odinger Equation. However, even accurate approximations of solving Schr¨odinger Equation, such as high order coupled cluster theories, require computational efforts that are too demanding to be applied on large chemical systems. This thesis tackles the problem of curse of dimensionality: how to reduce the time complexity of high-accuracy coupled cluster methods in order to accelerate computations of molecular energy. On one hand, we believe that low-rank approximation (i.e. Tensor HyperContraction) of high-order tensors appearing in coupled cluster theory is a promising way to achieve rank-reduced coupled cluster theory. On …