Open Access. Powered by Scholars. Published by Universities.®
Physical Sciences and Mathematics Commons™
Open Access. Powered by Scholars. Published by Universities.®
- Discipline
-
- Oceanography and Atmospheric Sciences and Meteorology (746)
- Oceanography (692)
- Life Sciences (588)
- Fresh Water Studies (364)
- Physics (361)
-
- Marine Biology (333)
- Environmental Sciences (271)
- Chemistry (259)
- Computer Sciences (224)
- Earth Sciences (146)
- Engineering (133)
- Animal Sciences (125)
- Ecology and Evolutionary Biology (114)
- Condensed Matter Physics (98)
- Zoology (93)
- Organic Chemistry (86)
- Natural Resources and Conservation (68)
- Social and Behavioral Sciences (65)
- Polymer Chemistry (63)
- Biodiversity (51)
- Geology (51)
- Atomic, Molecular and Optical Physics (49)
- Natural Resources Management and Policy (47)
- Materials Science and Engineering (39)
- Mathematics (37)
- Analytical Chemistry (36)
- Biochemistry, Biophysics, and Structural Biology (36)
- Plasma and Beam Physics (36)
- Physical Chemistry (34)
- Keyword
-
- Catadioptric imaging (1)
- Control (1)
- Criminal Network (1)
- GIS (1)
- Infectious Disease (1)
-
- Infrared surveillance (1)
- Integer Programming (1)
- Lamb waves (1)
- Locating-Dominating (1)
- Machine learning (1)
- Marsh (1)
- Mathematical Modeling (1)
- Modeling (1)
- Multiscale Models (1)
- Network Location (1)
- Optimization (1)
- Oyster reef (1)
- Pattern formation (1)
- Spatial analysis (1)
- Ultrasonic communication (1)
- Publication Year
Articles 1 - 30 of 1876
Full-Text Articles in Physical Sciences and Mathematics
Mid-21st Century Chesapeake Bay Hypoxia: The Role Of Climate Change At The Atmospheric, Terrestrial, And Oceanic Boundaries, Colin Hawes
Dissertations, Theses, and Masters Projects
Climate change is already increasing the volume and duration of coastal hypoxia and threatening living resources, particularly in eutrophic ecosystems. Atmospheric warming exacerbates deoxygenation by decreasing gas solubility and enhancing respiration and remineralization. Changes in terrestrial runoff and sea level influence hypoxia via nutrient availability and altered water temperature, respectively. However, the impacts of other future climate changes, including winds, shortwave and longwave radiation, non-runoff precipitation, and ocean water conditions, are still unknown. In this study, the impacts of such future climate changes on hypoxia in Chesapeake Bay were examined using a 3-D coupled estuarine hydrodynamic–biogeochemical model linked to a …
Low-Rank Matrix And Tensor Models For Data Science Applications, Jeremy Moulton Myers
Low-Rank Matrix And Tensor Models For Data Science Applications, Jeremy Moulton Myers
Dissertations, Theses, and Masters Projects
Low-rank approximations play an important role in data science analysis and applications. When the model is linear and the data is represented as a matrix, the optimal rank-r approximation is given by the r dominant factors of the singular value decomposition (SVD). When the model is multilinear and the data is represented as a multi-way array called a tensor, the optimal rank-r approximation may not even exist. Nonetheless, the canonical polyadic decomposition (CP) provides a useful low-rank tensor approximation for interpretability and analysis in a way similar to the matrix SVD but across multiple axes simultaneously. In this dissertation, we …
Environmental Controls On Carbon And Nitrogen Cycling In Alaskan Arctic Coastal Lagoons, Brian Seung Taek Kim
Environmental Controls On Carbon And Nitrogen Cycling In Alaskan Arctic Coastal Lagoons, Brian Seung Taek Kim
Dissertations, Theses, and Masters Projects
At the intersection of land, sea, and atmosphere, coastal lagoons act as bioreactors, processing both terrestrial and autochthonous organic matter production before export to the coastal ocean. Approximately half of the Beaufort Sea coast is outlined by barrier island chains that enclose shallow lagoons and sounds. These lagoons are subject to extreme seasonal variations in ice cover, temperature, and salinity, yet are home to a diverse and productive food web. Bound between the arctic tundra and Beaufort Sea, these shallow systems receive and process resources from both sea and particularly land in the form of coastal erosion and riverine export. …
Artificial Illumination Of Trawl Gear Components To Reduce Pacific Halibut (Hippoglossus Stenolepis) Bycatch In The U.S. West Coast Groundfish Bottom Trawl Fishery, Derek Jackson
Dissertations, Theses, and Masters Projects
Pacific halibut (Hippoglossus stenolepis) is a prohibited species for the U.S. West Coast Bottom Trawl Fishery and in the last decade, there has been a concentrated interest in the use of artificial illumination serving as a potential bycatch reduction device. Previous studies conducted off the coast of Oregon have found that the addition of green light-emitting diodes to the bridles of low-rise, cutback trawls greatly reduced the number of Pacific halibut caught. However, recent regulation changes now permit high-rise trawls, a gear configuration that fishes a very different volume of water than the previously permissible gear profile, in areas where …
Policy Recommendations For Tire Additive 6ppd And Its Derivative 6ppd-Q, Ashley E. King
Policy Recommendations For Tire Additive 6ppd And Its Derivative 6ppd-Q, Ashley E. King
Dissertations, Theses, and Masters Projects
Around 3.1 billion tires are produced around the world annually1. The antioxidant additive, 6PPD (i.e., N-(1,3-dimethylbutyl)-N’-phenyl-p-phenylenediamine) is widely employed in passenger and commercial vehicle tires at 0.4-2% by mass to impede tire degradation2. Antioxidants are intended to migrate to tire surfaces and form protective films to prevent rubber oxidation. 6PPD is designed to react with oxidant species like ozone, intentionally forming chemical transformation products that can then escape from the tire and into the environment. 6PPD-Q (i.e., 2-anilino-5-[(4-methylpentan-2-yl)amino]cyclohexa-2,5-diene-1,4-dione) is one such transformation product.
After release and disbursement in the environment, 6PPD-Q is bioavailable to aquatic animals and mammals and acute …
Experimental Studies Of Neutral Particles And The Isotope Effect In The Edge Of Tokamak Plasmas, Ryan Chaban
Experimental Studies Of Neutral Particles And The Isotope Effect In The Edge Of Tokamak Plasmas, Ryan Chaban
Dissertations, Theses, and Masters Projects
The H-mode plasma edge is a region of steep gradients in density and temperature known as the “pedestal” which greatly increases energy confinement. The complex links between neutral-plasma interactions and both diffusive and convective transport in the pedestal must be understood to model, predict, and achieve the high performance required for a fusion power plant. This dissertation explores the effects of different hydrogenic isotope neutral particles and plasma transport from the edge pedestal region into the Scrape-Off Layer. Current experiments typically use deuterium (H with amu=2 or D), however future fusion power plants may startup with hydrogen (H), and eventually …
Exploring Transient Execution Vulnerabilities, Side-Channel Attacks, And Defenses, Tao Zhang
Exploring Transient Execution Vulnerabilities, Side-Channel Attacks, And Defenses, Tao Zhang
Dissertations, Theses, and Masters Projects
Modern microprocessors utilize branch prediction and speculative execution to enhance instruction throughput. Instead of stalling the pipeline and waiting for branch targets to be computed, the CPU consults branch predictors for a possible destination and performs speculative execution. These microarchitectural techniques improve the efficiency of instruction pipelining and out-of-order execution, enabling higher performance and better resource utilization. Despite their widespread adoption, the potential security implications of branch misprediction and transient execution have not drawn much attention until recently. Around early 2018, the discovery of Spectre attacks exposed critical vulnerabilities in CPUs, undermining both software and hardware isolation and confidentiality. These …
Projected Impacts Of Climate Change And Watershed Management On Carbonate Chemistry And Oyster Growth In A Coastal Plain Estuary, Catherine Czajka
Projected Impacts Of Climate Change And Watershed Management On Carbonate Chemistry And Oyster Growth In A Coastal Plain Estuary, Catherine Czajka
Dissertations, Theses, and Masters Projects
Coastal acidification, warming, and nutrient management actions all alter water quality conditionsthat marine species experience, with potential impacts to their physiological processes. Decreases in calcite saturation state (ΩCa) and food availability, combined with warming water temperatures, pose a threat to calcifying organisms; however, the magnitude of future changes in estuarine systems is challenging to predict and not well known. This study aims to determine how and where oysters will be affected by future acidification, warming, and nutrient reductions, and the relative effects of these stressors. To address these goals, an oyster bioenergetics model for Eastern oysters (Crassostrea virginica) was embedded …
Learning-Based Sensing For Solving Health-Related Problems, Minglong Sun
Learning-Based Sensing For Solving Health-Related Problems, Minglong Sun
Dissertations, Theses, and Masters Projects
In the field of ubiquitous computing, health-related problem analysis has gained increasing attention. Collaborations between domain doctors and computing researchers have been established to recognize and address health-related issues. However, accurate detection and recognition of health-related problems remain a major challenge that attracts extensive research efforts. Among all the research works, wearable sensors-based health-related problem recognition is promising as it is low cost, low power, and easy to carry. This dissertation focuses on utilizing wearable sensors to study health-related problems. The first project introduced in this dissertation is TremorSense, a PD tremor detection system designed to classify Parkinson's Disease hand …
Extensions Of The Standard Model With Improved Ultraviolet Behavior, Mikayla Anderson
Extensions Of The Standard Model With Improved Ultraviolet Behavior, Mikayla Anderson
Dissertations, Theses, and Masters Projects
Although general relativity and the standard model have proved incredibly consistent at all scales accessible to tests, they are not expected to accurately describe nature at all scales; we know there is new physics to be discovered at higher energy scales (shorter distances). The nonrenormalizability of gravity prohibits a predictive quantum field theory description, unless the infinite parameter space needed to absorb divergences can be constrained. An asymptotically safe theory is one in which all of the couplings in the theory run to either zero or a nonzero ultraviolet fixed point. Requiring that a coupling reach an ultraviolet fixed point …
Scheduled Contrastive Loss In Continued Transfer Learning For Software Engineering Tasks, Aaron Michael Harris
Scheduled Contrastive Loss In Continued Transfer Learning For Software Engineering Tasks, Aaron Michael Harris
Dissertations, Theses, and Masters Projects
Considerable research has been performed with regard to using text-to-text machine learning methods to perform various software engineering tasks. At the same time, contrastive learning has shown promise in other modalities, such as computer vision-related problems, and has been explored to some extent in terms of limited software engineering tasks. We demonstrate that contrastive loss, on its own, is insufficient to surpass current baselines for these tasks; however, we note that there is a high degree of orthogonality in the results from existing and contrastive models. We show that when our contrastive method is used as an additional transfer learning …
Emergent Capabilities Of Llms For Software Engineering, Conor O'Brien
Emergent Capabilities Of Llms For Software Engineering, Conor O'Brien
Dissertations, Theses, and Masters Projects
A growing interest for Large Language Models (LLMs) is how increasing their size might result in changes to their behavior not predictable from relatively smaller-scaled models. Analyzing these emergent capabilities is therefore crucial to understanding and developing LLMs. Yet, whether LLMs exhibit emergence, or possess emergent capabilities, is a contested question. Furthermore, most research into LLM emergence has focused on natural language processing tasks and models suited for them. We focus on investigating emergence in the context of software engineering, and recontextualize the discussion of emergence in the context of prior research. We propose a multifaceted pipeline for evaluating and …
Automated Bug Report Management To Enhance Software Development, Yang Song
Automated Bug Report Management To Enhance Software Development, Yang Song
Dissertations, Theses, and Masters Projects
Bug report management is crucial yet challenging process that affects the efficiency of software development process. It involves reporting, triaging, detecting duplicates, assigning, localizing, fixing bugs, and thorough verification. The high volume and variety of bug reports complicate these tasks, highlighting the need for innovative solutions to improve the process and boost development efficiency. This dissertation explores the potential of automating the bug management process to optimize the effectiveness of software development and maintenance. It focuses on three key stages of bug management: reporting, assignment, and localization, presenting four innovative solutions for these phases. First, it discusses the challenges faced …
Investigating The Latent Security And Privacy Risks In Consumer-Oriented Software Systems, Kaushal Kafle
Investigating The Latent Security And Privacy Risks In Consumer-Oriented Software Systems, Kaushal Kafle
Dissertations, Theses, and Masters Projects
Consumer-oriented software systems have become the foundation on which consumer data is collected and transported from the consumers to the data processors. They are complex, with various interconnected, heterogeneous components working together, making their security and privacy analysis challenging, and impact on the user uncertain. In this work, we first explore how security threats can arise in novel context in such systems by performing a security evaluation of data-store based smart home platforms and the overall security risks posed by the design of routines within such platforms. We analyze various components of the smart home such as the platform’s permission …
Transitions To Two-Hadron States From Quantum Chromodynamics, Felipe Ortega Gama
Transitions To Two-Hadron States From Quantum Chromodynamics, Felipe Ortega Gama
Dissertations, Theses, and Masters Projects
The rich spectrum of hadrons reflects the complexity of interactions between quarks and gluons confined within them. Most of these hadrons are extremely short-lived and are called resonances. Experimentally, they are observed indirectly through their effects on the energy distribution in scattering experiments. Additionally, the non-perturbative nature of Quantum Chromodynamics (QCD), which governs the dynamics of quarks and gluons, prevents the implementation of known analytical techniques for calculating transition and interaction rates between hadrons. Lattice QCD (LQCD), a numerical implementation of QCD, provides a non-perturbative approach to studying the spectrum, as long as we understand how to account for finite-volume …
Probing With Displacements For Variance Reduction And The Effectiveness Of Sketched Krylov Eigenvalue Solvers, Heather Maria Switzer
Probing With Displacements For Variance Reduction And The Effectiveness Of Sketched Krylov Eigenvalue Solvers, Heather Maria Switzer
Dissertations, Theses, and Masters Projects
Scientific Computing is a multidisciplinary field that intersects Computer Science, Mathematics, and some other discipline to address complex problems utilizing computational systems. Numerical Linear Algebra (NLA), a subfield within Scientific Computing, develops and analyzes numerical algorithms for tasks involving linear operators such as matrices and their transformations. This dissertation focuses on developing kernels for two specific areas of NLA. Firstly, we explore variance reduction methods for trace approximation. In Lattice Quantum Chromodynamics (LQCD), computing the trace of a matrix inverse is crucial for investigating interactions among quarks and gluons in subatomic space. However, directly computing a matrix inverse is computationally …
A Reevaluation Of Why Crypto-Detectors Fail: A Systematic Revaluation Of Cryptographic Misuse Detection Techniques, Scott Marsden
A Reevaluation Of Why Crypto-Detectors Fail: A Systematic Revaluation Of Cryptographic Misuse Detection Techniques, Scott Marsden
Dissertations, Theses, and Masters Projects
The correct use of cryptography is central to ensuring data security in modern software systems. Hence, several academic and commercial static analysis tools have been developed for detecting and mitigating crypto-API misuse. While developers are optimistically adopting these crypto-API misuse detectors (or crypto-detectors) in their software development cycles, this momentum must be accompanied by a rigorous understanding of their effectiveness at finding crypto-API misuse in practice. The original paper presents the MASC framework, which enables a systematic and data-driven evaluation of crypto-detectors using mutation testing. MASC was grounded in a comprehensive view of the problem space by developing a data-driven …
Exploring Software Licensing Issues Faced By Legal Practitioners, Nathan James Wintersgill
Exploring Software Licensing Issues Faced By Legal Practitioners, Nathan James Wintersgill
Dissertations, Theses, and Masters Projects
Most modern software products incorporate open source components, which requires compliance with each component’s licenses. As noncompliance can lead to significant repercussions, organizations often seek advice from legal practitioners to maintain license compliance, address licensing issues, and manage the risks of noncompliance. While legal practitioners play a critical role in the process, little is known in the software engineering community about their experiences within the open source license compliance ecosystem. To fill this knowledge gap, a joint team of software engineering and legal researchers designed and conducted a survey with 30 legal practitioners and related occupations and then held 16 …
Impacts And Uncertainties Of Climate Change On The Chesapeake Bay, Kyle E. Hinson
Impacts And Uncertainties Of Climate Change On The Chesapeake Bay, Kyle E. Hinson
Dissertations, Theses, and Masters Projects
Climate change impacts in the Chesapeake Bay will limit the efficacy of nutrient reduction efforts and decrease dissolved oxygen, but uncertainties associated with the magnitude of these effects remain. An understanding of underlying mechanisms that have driven recent warming trends will narrow uncertainties for future pathways of temperature change. Additionally, future simulations of climate impacts in the estuary are dependent on multiple different sources of uncertainty, many of which have not yet been fully evaluated. This dissertation used a three- dimensional coupled hydrodynamic-biogeochemical model to investigate recent warming trends as well as underlying uncertainties likely to influence regional projections of …
Succession Of The Late Summer Phytoplankton Blooms In The York River Estuary, Va, Heather Kathleen Corson
Succession Of The Late Summer Phytoplankton Blooms In The York River Estuary, Va, Heather Kathleen Corson
Dissertations, Theses, and Masters Projects
The influence of bottom-up and top-down controls on the formation and persistence of phytoplankton blooms has been well studied. However, the relative importance of these bottom-up and top-down controls vary spatially and temporally. In the tidal tributaries and mainstem of Chesapeake Bay, the summer dinoflagellate population follows a succession of bloom-producing species. The dinoflagellate species Margalefidinium polykrikoides and Alexandrium monilatum are currently considered the end of this succession. These species form near-annual blooms in the lower half of Chesapeake Bay and are considered harmful algal bloom (HAB) species due to their negative ecological impacts. However, analysis of long-term monitoring data …
Investigating The Environmental And Molecular Dependence Of Single-Molecule Blinking-Based Multiplexing, Grace Desalvo
Investigating The Environmental And Molecular Dependence Of Single-Molecule Blinking-Based Multiplexing, Grace Desalvo
Dissertations, Theses, and Masters Projects
Visualizing nanoscale biological systems allows us to uncover their detailed structure and functions, which have major implications in the biomedical field. Super-resolution microscopy is a powerful tool for fluorescence imaging because, by overcoming the diffraction limit of light, it accesses structural detail with unprecedented spatial resolution. Although multicolor super-resolution imaging has been successfully implemented in many experiments, its efficiency is limited by reliance on spectral measurement for emitter identification, which limits the combinations of compatible probes to be used together. Blinking-based multiplexing (BBM) is a novel approach that circumvents the need for spectrally-distinct emitters by instead exploiting the intrinsic differences …
A Comprehensive Study Of Bills Of Materials For Software Systems, Trevor Stalnaker
A Comprehensive Study Of Bills Of Materials For Software Systems, Trevor Stalnaker
Dissertations, Theses, and Masters Projects
Software Bills of Materials (SBOMs) have emerged as tools to facilitate the management of software dependencies, vulnerabilities, licenses, and the supply chain. Significant effort has been devoted to increasing SBOM awareness and developing SBOM formats and tools. Despite this effort, recent studies have shown that SBOMs are still an early technology not adequately adopted in practice yet, mainly due to limited SBOM tooling and lack of industry consensus on SBOM content, tool usage, and practical benefits. Expanding on previous research, this paper reports a comprehensive study that first investigates the current challenges stakeholders encounter when creating and using SBOMs. The …
From Field To Shore: Policy Improvements To Better Protect Ecosystem And Human Health From Poultry Wastes In Virginia, Rachael A. King
From Field To Shore: Policy Improvements To Better Protect Ecosystem And Human Health From Poultry Wastes In Virginia, Rachael A. King
Dissertations, Theses, and Masters Projects
Commercial poultry production is a large industry with economic importance in Virginia. However, mismanagement of manures and wastes generated from the poultry industry can be a serious threat to ecosystem and human health. Primarily, there is a concern about nutrient pollution related to runoff and infiltration of poultry wastes. High nutrient loading can cause groundwater contamination, eutrophication, and harmful algal blooms (HABs). Additional concerns include metals, antibiotics, and pesticides that can be found in poultry wastes. To determine existing threats to ecosystem and human health from the poultry industry, Virginia policies regulating the industry are reviewed and their strengths and …
Ecosystem Transitions And State Changes Rapidly Alter The Coastal Carbon Landscape: Evidence From The Chesapeake Bay Region, Alexander Jason Smith
Ecosystem Transitions And State Changes Rapidly Alter The Coastal Carbon Landscape: Evidence From The Chesapeake Bay Region, Alexander Jason Smith
Dissertations, Theses, and Masters Projects
The coastal landscape is a naturally shifting mosaic of distinct ecosystems that are rapidly migrating with climate change. While directional changes in climate, such as warming and sea level rise, are fundamentally reorganizing the coastal landscape, ecosystem function, especially carbon storage, is affected to an unknown degree. This dissertation presents four chapters that examine the role of ecosystem transitions in coastal carbon dynamics across a range of spatial scales – within individual ecosystems, between two ecosystems, and at the landscape between an array of ecosystems. Ghost forests, or the marsh-forest ecotone, serves as an ideal example of a migratory ecotone. …
Gas-Phase Proton Affinities Of Proline- And Pipecolic Acid-Containing Dipeptides, Trinh Ton
Gas-Phase Proton Affinities Of Proline- And Pipecolic Acid-Containing Dipeptides, Trinh Ton
Dissertations, Theses, and Masters Projects
Mass spectrometry (MS) is one of the most used techniques in proteomics because it allows for both high-throughput and quantitative analyses. Bottom-up MS-based proteomics involves breaking down proteins into smaller chains of amino acids called peptides, ionizing and fragmenting the peptides, and identifying the fragments using sequencing databases. These databases depend on the random fragmentation at the backbone peptide bond of the peptides, as predicted by the mobile proton model. Research has shown that peptides containing proline or pipecolic acid have selective fragmentations that could lead to incorrect identification in the sequencing algorithms. These selective cleavages are called “the proline …
Chesapeake Bay Carbonate Cycle: Past, Present, And Future, Fei Da
Chesapeake Bay Carbonate Cycle: Past, Present, And Future, Fei Da
Dissertations, Theses, and Masters Projects
Multiple natural and anthropogenic drivers are expanding the variability of the estuarine carbonate system (CO2 system). These changes in the CO2 system are threatening the health of ecologically and economically important bivalve species. This dissertation investigates the Chesapeake Bay CO2 system by using numerical models and historical water quality data, focusing on the past three decades, the contemporary period, and the late 2060s. In Chapter 2, sensitivity experiments are conducted with a 3-D Chesapeake Bay hydrodynamic-biogeochemical model and reveal that the magnitude of decadal trends in the CO2 system over the past 30 years is much greater than that observed …
Domain-Specific Optimization For Machine Learning System, Yu Chen
Domain-Specific Optimization For Machine Learning System, Yu Chen
Dissertations, Theses, and Masters Projects
The machine learning (ML) system has been an indispensable part of the ML ecosystem in recent years. The rapid growth of ML brings new system challenges such as the need of handling more large-scale data and computation, the requirements for higher execution performance, and lower resource usage, stimulating the demand for improving ML system. General-purpose system optimization is widely used but brings limited benefits because ML applications vary in execution behaviors based on their algorithms, input data, and configurations. It's difficult to perform comprehensive ML system optimizations without application specific information. Therefore, domain-specific optimization, a method that optimizes particular types …
Tracing Atlantic Sea Scallops Using Radio Frequency Identification (Rfid) Technology, Will Shoup
Tracing Atlantic Sea Scallops Using Radio Frequency Identification (Rfid) Technology, Will Shoup
Dissertations, Theses, and Masters Projects
Traceable seafood can be linked back to its origin and method of catch. Improving the traceability of marine organisms involves establishing a transparent Chain of Custody (CoC) by collecting data at checkpoints throughout the supply chain, from ship to shore to store. This report explores the feasibility of integrating Radio Frequency Identification (RFID) technology into the United States Atlantic sea scallop (Placopecten magellanicus) fishery in order to improve traceability. This report serves as a forward-looking evaluation of RFID technology that is intended to inform interested stakeholders of its functionality and capabilities. It is not intended to serve as a management …
Mattanock Town Restoration Plan, Katlin Mccarter Grigsby
Mattanock Town Restoration Plan, Katlin Mccarter Grigsby
Dissertations, Theses, and Masters Projects
Mattanock Town's Restoration Plan is a science-based restoration process that evaluates the site's history, the tribal history, and the most current research to maximize native habitats, enhance coastal resilience, and reconnect the Nansemond people to the local river. Restoration priorities include increasing native plant species, incorporating oyster habitat, and addressing erosion. This plan details how synthesizing existing and new physical, biological, and cultural information can help the Nansemond Indian Nation prioritize projects that benefit their community and the surrounding environment.
Recoverable Memory Bank For Class-Incremental Learning, Jiangtao Kong
Recoverable Memory Bank For Class-Incremental Learning, Jiangtao Kong
Dissertations, Theses, and Masters Projects
Incremental learning aims to enable machine learning systems to sequentially learn new tasks without forgetting the old ones. While some existing methods, such as data replay-based and parameter isolation-based approaches, achieve remarkable results in incremental learning, they often suffer from memory limits, privacy issues, or generation instability. To address these problems, we propose Recoverable Memory Bank (RMB), a novel non-exemplar-based approach for class incremental learning (CIL). Specifically, we design a dynamic memory bank that stores only one aggregated memory representing each class of the old tasks. Next, we propose a novel method that combines a high-dimensional space rotation matrix and …