-
Prospective Students
-
Faculty and Staff
-
Current Students
-
Alumni and Friends
The development of in-memory computing has fueled the emergence of in-memory computing systems. Data explosion is also posing an unprecedented demand for memory capacity to handle the ever-growing data size. Thus, in-memory computing systems are increasingly looking inward at hybrid memory caches of under-processed data as resources to be mined. Our preliminary study finds that some existing data management strategies often trade application performance for low memory utilization, and hence can induce frequent I/O operations between memory system and storage system.
To achieve this goal, we propose to design a hybrid memory system that includes fast and relatively slow memory hardware. In order to realize a runtime system that automatically optimizes data management on hybrid memory, we will (1) propose a new shared in-memory cache layer among parallel executors that are co-hosted on the same computing node, which aims to improve the overall hit rate of data blocks; (2) develop a middleware layer built on top of existing deep learning frameworks that streamlines the support and implementation of online learning applications; (3) design a unified in-memory computing architecture with efficient data management strategy to optimize memory allocation and recycle for ML applications.
The metropolitan areas suffer more traffic, the change in travel time is very complex as it can be influenced by various factors, many of which are also unpredictable. Random forest was applied in the travel time prediction application to overcome the overfitting problem. Furthermore, the attention mechanism is implemented by developing the neural network to capture the inner relationship within the traffic data. The proposed long short memory neural network with attention mechanism method achieves its superior capability for TTP longer than 15 minutes (30 min to 60 min), overcoming the performance issue through long temporal dependency and memory blocks. To validate the accuracy and reliability of proposed models, the proposed approaches are tested using a freeway corridor in Charlotte, North Carolina, using the probe vehicle-based traffic data. Detailed information about the input variables and data preprocessing was presented. The results indicate that all proposed TTP models predicting in 15 minutes show better prediction performance over the other time horizons. A comparison with other prediction methods validates that the proposed hybrid LSTM and RF method can achieve a better prediction performance in accuracy and efficiency, proving its deployment is one of the successful solutions to critical, real-world transportation challenges.
The antibiotic resistance genes (ARGs) have been increasing over time in the environment due to human activities and antibiotic use. According to CDC, antibiotic resistance (AR) causes casual infections untreatable and result in high socioeconomic costs and health care burdens. This study focuses on targeting ARGs origination, distribution, and expression in the water system under anthropogenic effects. Investigation of ARGs takes following goals: 1) Optimization of conventional and new technologies for ARG detection using quantitative polymerase chain reaction (qPCR) and droplet digital PCR (qPCR), 2) Comparison of performance between qPCR and ddPCR on ARGs, 3) Measurement of ARG abundance throughout different types of water bodies such as a lake, river, wetland, underground aquifer across the U.S. water system under anthropogenic effects, and 4) metagenomic bacterial and ARG expression analysis under a stressed environment with metal-laden industrial flue gas desulfurization (FGD) wastewater. This study helps improve surveillance and develop mitigation plans for AR and ARGs in the water system globally. And add knowledge to have better control in tracking, treatment, and containment protecting the health of humans and the ecosystem.
Courts of last resort in the United States are becoming increasingly important in American politics as the number of cases, influential decisions, and controversial issues continue to rise in the states. In discussions of federalism in the United States, these critical institutions are often overlooked as a complex system, often due to substantial data limitations on the behavior and outcomes of these courts. I situate state courts of last resort as a complex adaptive system in the broader US framework. I then seek to redress the data shortcomings by introducing a comprehensive database on state courts of last resort from 1953-2010. Using advanced data capture techniques, I evaluate my parsers to capture the ever-changing structures of the source documents. This database will be the largest in scope and case detail to date. Moreover, it should further our understanding of judicial decision making and assist the prediction of the impact of institutional change on the system.
In addition, I modeled and analyzed the system as a complex adaptive system. Since the system has network characteristics, I used the approach of network science to model the system based on the citation behavior. Moreover, I created an automated dictionary-based classification model to extract and classify the citation treatments for the court cases.
Using state-of-the-art algorithms in network science and natural language processing, I was able to analyze the system and test the performance of the algorithms based on the system characteristics.
The modern economy runs with heavy reliance on the free flow of goods across the international logistic and supply chain. Advances in international freight transportation systems supported by intermodal integration, freight containerization, hub-and-spoke shipping system, and supply chain security, has reduced the distance friction of flow of goods and drastically lowered physical barriers of commercial activities. However, it is little known yet how spatial interactions of trade and shipping take place under the complex logistic chain process and what spatial phenomena ensue from such processes. In this dissertation, I study the nature of spatial interaction phenomena in the context of the contemporary state of the international transportation system. First, I study how the spatial structure of the port system is formed with intermodal integration of the modern international logistic system across land and water. Second, I explore how the hub-and-spoke system in the international transportation network contributes to the global shrinkage of space. Third, I investigate the effect of domestic armed conflicts developed by political instability on freight mobility and ensuing differential openness of regions to the global market. Results of the three pieces of research are as follows. First, the spatial structure of the port system is found to comprise interdependent collections of hinterlands, feeder and hub ports, and forelands along a logistical continuum, which mirror the functional division of logistic processes across space. Second, the hub-and-spoke shipping system reduces the distance friction of shipping flows and is the main driver of global shrinkage of space in terms of long-distance trade. Third, freight mobility is found to be greatly compromised by the lack of logistic chain security stemming from prevailing armed violence along inland transportation corridors. The findings confirm that intermodal logistic integration, hub-and-spoke distribution system and supply chain security are important key components of the modern international transportation system that determine global-scale spatial organization, shipping flow and freight mobility.
Renewable energy-based electric power generation is receiving more attention due to the increasing power demand and environmental concerns. The interconnection of these distributed resources to the grid is based on the grid code standards to ensure power quality, reliability, and security. The intermittent nature of the renewable penetrated grid, along with phasing out of conventional generation units, new HVDC lines, and long-distance transmissions from remote areas, impose several challenges to grid stability. Hence it is crucial to explore control approaches that can efficiently control these Distributed Energy Resources (DERs) to improve the overall power quality and reliability. This dissertation presents modeling, stability studies, and advanced control architectures that can support and coordinate the Wind Energy Conversion Systems (WECS) and other inverter-based Distributed Energy Resources (DERs) to improve the quality of the generation, transmission, and distribution systems. The first part of the work proposes advanced adaptive-based robust sensorless control approaches for rotor side and grid side control of DFIG based WECS.
Further, the dissertation discusses the challenges of transferring high power of renewable penetrated grid and some possible solutions and control approaches. Finally, the work efficiently coordinates the available resources to ensure power quality in a distribution network. All the proposed designs are validated using simulation results developed by dynamic models or through real-time simulators, which proves the ability of the advanced controllers to improve grid reliability. The quantification based on standard metrics used for performance improvements discussed in each design shows that the designs have exceptional advantages compared with conventional controllers.
The overwhelming number of recent data breaches reported hundreds of terabytes of highly sensitive information, including national, financial, and personal, have been stolen from different organizations, indicating clear asymmetric disadvantage defender faces against cyber attackers. Modern attackers are well organized, highly stealthy, and stay persistent in the network for years; therefore, known as an advanced persistent threat (APT). Existing detection and prevention based cyber defense techniques usually approach the target for specific, known attack signatures, descriptions, and behaviors. However, APT attackers can easily avoid such detection techniques employing reconnaissance, fingerprinting, and social engineering. It is often very challenging and sometimes infeasible for defenders to prevent the information gathering of the adversary and patch all the vulnerabilities in the system. Therefore, a proactive defense approach is needed to break such asymmetry.
Active Cyber Defense (ACD) is a promising paradigm to achieve this goal. ACD can proactively mislead adversaries and enables a unique opportunity to engage with them to learn new attack tactics and techniques. ACD enhances real-time detection, analysis, and mitigation of APT attacks. ACD can be achieved through cyber agility and cyber deception. Cyber Agility, such as moving target defense (MTD), enables cyber systems to defend proactively against sophisticated attacks by dynamically changing the system configuration parameters (called mutable parameters) in order to deter adversaries from reaching their goals. On the other hand, Cyber Deception is an intentional misrepresentation of the system's ground truth to manipulate adversaries' actions.
Although cyber deception and MTD have been around for more than decades, static configurations and the lack of automation made many of the existing techniques easily discoverable by attackers and too expensive to manage, which diminishes the value of these technologies. Sophisticated APTs are very dynamic and thereby require a highly adaptive and embedded defense that can dynamically create honey resources and orchestrate the ACD environment appropriately according to the adversary behavior in real-time.
To overcome these challenges, this dissertation introduced an autonomous resilient ACD framework, having the following aspects: (1) developing multistrategy ACD policies that leverage an optimal dynamic composition of various MTD and deception techniques to maximize the defense utility, (2) a policy specification language and an extensible rich API integrated with a synthesis engine for developing different MTD techniques without consulting about the low-level network and system configuration management, (3) a theoretical framework and implementation for an autonomous goal-oriented cyber deception planner that optimizes deception decision-making.
Structured light systems (SLS) have become increasingly important for three-dimensional shape measurements. A quantitative evaluation of the spatial resolution is also becoming increasingly important. The spatial frequency response of the instrument is a reasonable metric for resolution and is commonly referred to as the instrument transfer function (ITF). We used the ITF to determine the capability of a commercial SLS the EinScan-pro using a step artifact. The ITF is similar to the modulation transfer function (MTF) of an imaging system, which describes how well the system images an object as a function of spatial frequency. Similarly, ITF describes the instrument response to the spatial frequency on the surface to be measured. Many optical measurement instruments use a camera for data acquisition and the optical transfer function will necessarily impose a limit on the instrument resolution. The ITF and the MTF metrics rely on the linearity of the measurement. Only a liner and shift invariant system can be used to uniquely define ITF/MTF. In this dissertation, we describe the use of the step artifact to determine the spatial resolution of a commercial SLS the EinScan-Pro. We check the use of the ITF over the MTF of the imaging system. We present a methodology to check the combined uncertainty for the ITF measurements including a method to check the applicability of the step artifact for the ITF measurements, the impact of the use of step artifact with different surface finishes, and the effect of the tilt and disposition of the artifact during the measurement. We also, present the use of the bispectrum for the non-linearity check of any kind of measurement which is applicable for both ITF/MTF measurements.
We propose methods for functional predictor selection and the estimation of smooth functional coefficients simultaneously in a scalar-on-function regression problem under a high-dimensional multivariate functional data setting. In particular, we develop two methods for functional group-sparse regression under a generic Hilbert space of infinite dimension. We show the convergence of algorithms and the consistency of the estimation and the selection (oracle property) under infinite-dimensional Hilbert spaces. Simulation studies show the effectiveness of the methods in both the selection and the estimation of functional coefficients. The applications to functional magnetic resonance imaging (fMRI) reveal the human brain regions related to ADHD and IQ. In addition, we apply the proposed methods to an econometric data set to find the related functional covariates to GDP of a country. To extend the results, we propose numerical algorithms for more complex models, such as nonlinear (via RKHS), logistic, sparse function--on--function, and standardization of the results of the sparse scalar--on--function models before we list the applications of these extensions to the brain image data analysis.
