Engineering Commons^™

An Efficient Privacy-Preserving Framework For Video Analytics, Tian Zhou

Doctoral Dissertations

With the proliferation of video content from surveillance cameras, social media, and live streaming services, the need for efficient video analytics has grown immensely. In recent years, machine learning based computer vision algorithms have shown great success in various video analytic tasks. Specifically, neural network models have dominated in visual tasks such as image and video classification, object recognition, object detection, and object tracking. However, compared with classic computer vision algorithms, machine learning based methods are usually much more compute-intensive. Powerful servers are required by many state-of-the-art machine learning models. With the development of cloud computing infrastructures, people are able …

Learning To Rig Characters, Zhan Xu

Doctoral Dissertations

With the emergence of 3D virtual worlds, 3D social media, and massive online games, the need for diverse, high-quality, animation-ready characters and avatars is greater than ever. To animate characters, artists hand-craft articulation structures, such as animation skeletons and part deformers, which require significant amount of manual and laborious interaction with 2D/3D modeling interfaces. This thesis presents deep learning methods that are able to significantly automate the process of character rigging. First, the thesis introduces RigNet, a method capable of predicting an animation skeleton for an input static 3D shape in the form of a polygon mesh. The predicted skeletons …

Design And Analysis Of Content Caching Systems, Anirudh Sabnis

Doctoral Dissertations

Caching is a simple yet powerful technique that has had a significant impact on improving the performance of various computer systems. From internet content delivery to CPUs, domain name systems, and database systems, caching has played a pivotal role in making these systems faster and more efficient. The basic idea behind caching is to store frequently accessed data locally, so that future requests for that data can be served more quickly. For example, a Content Delivery Network (CDN) like Akamai deploys thousands of edge caches across the globe, so that end-user requests can be served from a nearby cache, rather …

Anomaly Detection On Complex Health Information Technology Systems, Haoran Niu

Doctoral Dissertations

While modern complex computer systems provide enormous benefits to our daily lives, the increasing complexity of these large-scale systems also makes them more susceptible to unexpected software malfunctions and malicious attacks. This is especially true for Health Information Technology (HIT), which has revolutionized healthcare delivery by making it more efficient, effective, and accessible. Nevertheless, the widespread adoption of HIT has introduced new challenges related to ensuring system reliability and security. As a result, the development of novel algorithms and frameworks to detect anomalies in such systems has become increasingly important for enhancing patient safety and improving the efficiency and effectiveness …

Deep Learning Based Power System Stability Assessment For Reduced Wecc System, Yinfeng Zhao

Doctoral Dissertations

Power system stability is the ability of power system, for a giving initial operating condition, to reach a new operation condition with most of the system variables bounded in normal range after subjecting to a short or long disturbance. Traditional power system stability mainly uses time-domain simulation which is very time consuming and only appropriate for offline assessment.

Nowadays, with increasing penetration of inverter based renewable, large-scale distributed energy storage integration and operation uncertainty brought by weather and electricity market, system dynamic and operating condition is more dramatic, and traditional power system stability assessment based on scheduling may not be …

Development Of A Soft Robotic Approach For An Intra-Abdominal Wireless Laparoscopic Camera, Hui Liu

Doctoral Dissertations

In Single-Incision Laparoscopic Surgery (SILS), the Magnetic Anchoring and Guidance System (MAGS) arises as a promising technique to provide larger workspaces and field of vision for the laparoscopes, relief space for other instruments, and require fewer incisions. Inspired by MAGS, many concept designs related to fully insertable magnetically driven laparoscopes are developed and tested on the transabdominal operation. However, ignoring the tissue interaction and insertion procedure, most of the designs adopt rigid structures, which not only damage the patients' tissue with excess stress concentration and sliding motion but also require complicated operation for the insertion. Meanwhile, lacking state tracking of …

Toward Generating Efficient Deep Neural Networks, Chengcheng Li

Doctoral Dissertations

Recent advances in deep neural networks have led to tremendous applications in various tasks, such as object classification and detection, image synthesis, natural language processing, game playing, and biological imaging. However, deploying these pre-trained networks on resource-limited devices poses a challenge, as most state-of- the-art networks contain millions of parameters, making them cumbersome and slow in real-world applications. To address this problem, numerous network compression and acceleration approaches, also known as efficient deep neural networks or efficient deep learning, have been investigated, in terms of hardware and software (algorithms), training, and inference. The aim of this dissertation is to study …

Enabling Daily Tracking Of Individual’S Cognitive State With Eyewear, Soha Rostaminia

Doctoral Dissertations

Research studies show that sleep deprivation causes severe fatigue, impairs attention and decision making, and affects our emotional interpretation of events, which makes it a big threat to public safety, and mental and physical well-being. Hence, it would be most desired if we could continuously measure one’s drowsiness and fatigue level, their emotion while making decisions, and assess their sleep quality in order to provide personalized feedback or actionable behavioral suggestions to modulate sleep pattern and alertness levels with the aim of enhancing performance, well-being, and quality of life. While there have been decades of studies on wearable devices, we …

Data Scarcity In Event Analysis And Abusive Language Detection, Sheikh Muhammad Sarwar

Doctoral Dissertations

Lack of data is almost always the cause of the suboptimal performance of neural networks. Even though data scarce scenarios can be simulated for any task by assuming limited access to training data, we study two problem areas where data scarcity is a practical challenge: event analysis and abusive content detection} Journalists, social scientists and political scientists need to retrieve and analyze event mentions in unstructured text to compute useful statistical information to understand society. We claim that it is hard to specify information need about events using keyword-based representation and propose a Query by Example (QBE) setting for event …

Learning With Limited Labeled Data For Image And Video Understanding, Razieh Kaviani Baghbaderani

Doctoral Dissertations

Deep learning-based algorithms have remarkably improved the performance in many computer vision tasks. However, deep networks often demand a large-scale and carefully annotated dataset and sufficient sample coverage of every training category. However, it is not practical in many real-world applications where only a few examples may be available, or the data annotation is costly and require expert knowledge. To mitigate this issue, learning with limited data has gained considerable attention and is investigated thorough different learning methods, including few-shot learning, weakly/semi supervised learning, open-set learning, etc.

In this work, the classification problem is investigated under an open-world assumption to …

Task-Oriented Manipulation Planning: Teaching Robot Manipulators To Learn Trajectory Tasks, Yan Li

Doctoral Dissertations

As robot manipulator applications are conducted in more complex tasks and unstructured environments, traditional manual programming cannot match the growing requirements. However, human experts usually know how to operate robot manipulators to complete tasks, but they do not know how to manually program the robot for automatically executing tasks. From a general point of view, a robot manipulation task is composed of a series of consecutive robot actions and environment states which we call it trajectory task. Imitation learning, an emerging and popular technique of robot behavior programming, is a good way to tackle this line of work but still …

Data Parallel Frameworks For Training Machine Learning Models, Guoyi Zhao

Doctoral Dissertations

Machine learning is the study of computer algorithms that focuses on analyzing and interpreting patterns and structures in data. It has been successfully applied to many areas in computer science and achieved state-of-the-art results to enable learning, reasoning, and decision-making without human interactions. This research aims to develop innovated data parallel frameworks to accommodate the computing resources to parallelize different machine learning and deep learning algorithms and speed up the training. To achieve that, we explore three interesting frameworks in this dissertation: (1) Sync-on-the-fly framework for gradient descent algorithms on transient resources; (2) Asynchronous Proactive Data Parallel framework for both …

Models And Machine Learning Techniques For Improving The Planning And Operation Of Electricity Systems In Developing Regions, Santiago Correa Cardona

Doctoral Dissertations

The enormous innovation in computational intelligence has disrupted the traditional ways we solve the main problems of our society and allowed us to make more data-informed decisions. Energy systems and the ways we deliver electricity are not exceptions to this trend: cheap and pervasive sensing systems and new communication technologies have enabled the collection of large amounts of data that are being used to monitor and predict in real-time the behavior of this infrastructure. Bringing intelligence to the power grid creates many opportunities to integrate new renewable energy sources more efficiently, facilitate grid planning and expansion, improve reliability, optimize electricity …

Improving The Programmability Of Networked Energy Systems, Noman Bashir

Doctoral Dissertations

Global warming and climate change have underscored the need for designing sustainable energy systems. Sustainable energy systems, e.g., smart grids, green data centers, differ from the traditional systems in significant ways and present unique challenges to system designers and operators. First, intermittent renewable energy resources power these systems, which break the notion of infinite, reliable, and controllable power supply. Second, these systems come in varying sizes, spanning over large geographical regions. The control of these dispersed and diverse systems raises scalability challenges. Third, the performance modeling and fault detection in sustainable energy systems is still an active research area. Finally, …

Model Based Force Estimation And Stiffness Control For Continuum Robots, Vincent A. Aloi

Doctoral Dissertations

Continuum Robots are bio-inspired structures that mimic the motion of snakes, elephant trunks, octopus tentacles, etc. With good design, these robots can be naturally compliant and miniaturizable, which makes Continuum Robots ideal for traversing narrow complex environments. Their flexible design, however, prevents us from using traditional methods for controlling and estimating loading on rigid link robots.

In the first thrust of this research, we provided a novel stiffness control law that alters the behavior of an end effector during contact. This controller is applicable to any continuum robot where a method for sensing or estimating tip forces and pose exists. …

Distributed Learning Algorithms: Communication Efficiency And Error Resilience, Raj Kumar Maity

Doctoral Dissertations

In modern day machine learning applications such as self-driving cars, recommender systems, robotics, genetics etc., the size of the training data has grown to the point that it has become essential to design distributed learning algorithms. A general framework for the distributed learning is \emph{data parallelism} where the data is distributed among the \emph{worker machines} for parallel processing and computation to speed up learning. With billions of devices such as cellphones, computers etc., the data is inherently distributed and stored locally in the users' devices. Learning in this set up is popularly known as \emph{Federated Learning}. The speed-up due to …

Data-Driven Control, Modeling, And Forecasting For Residential Solar Power, Akansha Singh Bansal

Doctoral Dissertations

Distributed solar generation is rising rapidly due to a continuing decline in the cost of solar modules. Most residential solar deployments today are grid-tied, enabling them to draw power from the grid when their local demand exceeds solar generation and feed power into the grid when their local solar generation exceeds demand. The electric grid was not designed to support such decentralized and intermittent energy generation by millions of individual users. This dramatic increase in solar power is placing increasing stress on the grid, which must continue to balance its supply and demand despite the potential for large solar fluctuations. …

Federated Agentless Detection Of Endpoints Using Behavioral And Characteristic Modeling, Hansaka Angel Dias Edirisinghe Kodituwakku

Doctoral Dissertations

During the past two decades computer networks and security have evolved that, even though we use the same TCP/IP stack, network traffic behaviors and security needs have significantly changed. To secure modern computer networks, complete and accurate data must be gathered in a structured manner pertaining to the network and endpoint behavior. Security operations teams struggle to keep up with the ever-increasing number of devices and network attacks daily. Often the security aspect of networks gets managed reactively instead of providing proactive protection. Data collected at the backbone are becoming inadequate during security incidents. Incident response teams require data that …

Profile-Guided Data Management For Heterogeneous Memory Systems, Matthew B. Olson

Doctoral Dissertations

Market forces and technological constraints have led to a gap between CPU and memory performance that has widened for decades. While processor scaling has plateaued in recent years, this gap persists and is not expected to diminish for the foreseeable future. This discrepancy presents a host of challenges for scaling application performance, which have only been exacerbated in recent years, as increasing demands for fast and effective data analytics are driving memory energy, bandwidth, and capacity requirements to new heights.

To address these trends, hardware architects have introduced a plethora of memory technologies. For example, most modern memory systems include …

Qualitative And Quantitative Improvements For Positron Emission Tomography Using Different Motion Correction Methodologies, Tasmia Rahman Tumpa

Doctoral Dissertations

Positron Emission Tomography (PET) data suffers from low image quality and quantitative accuracy due to different kinds of motion of patients during imaging. Hardware-based motion correction is currently the standard; however, is limited by several constraints, the most important of which is retroactive data correction. Data-driven techniques to perform motion correction in this regard are active areas of research. The motivation behind this work lies in developing a complete data-driven approach to address both motion detection and correction. The work first presents an algorithm based on the positron emission particle tracking (PEPT) technique and makes use of time-of-flight (TOF) information …

On Improving Robustness Of Hardware Security Primitives And Resistance To Reverse Engineering Attacks, Vinay C. Patil

Doctoral Dissertations

The continued growth of information technology (IT) industry and proliferation of interconnected devices has aggravated the problem of ensuring security and necessitated the need for novel, robust solutions. Physically unclonable functions (PUFs) have emerged as promising secure hardware primitives that can utilize the disorder introduced during manufacturing process to generate unique keys. They can be utilized as \textit{lightweight} roots-of-trust for use in authentication and key generation systems. Unlike insecure non-volatile memory (NVM) based key storage systems, PUFs provide an advantage -- no party, including the manufacturer, should be able to replicate the physical disorder and thus, effectively clone the PUF. …

Resource Allocation In Distributed Service Networks, Nitish Kumar Panigrahy

Doctoral Dissertations

The past few years have witnessed significant growth in the use of distributed network analytics involving agile code, data and computational resources. In many such networked systems, for example, Internet of Things (IoT), a large number of smart devices, sensors, processing and storage resources are widely distributed in a geographic region. These devices and resources distributed over a physical space are collectively called a distributed service network. Efficient resource allocation in such high performance service networks remains one of the most critical problems. In this thesis, we model and optimize the allocation of resources in a distributed service network. This …

Cost-Efficient Resource Provisioning For Cloud-Enabled Schedulers, Lurdh Pradeep Reddy Ambati

Doctoral Dissertations

Since the last decade, public cloud platforms are rapidly becoming de-facto computing platform for our society. To support the wide range of users and their diverse applications, public cloud platforms started to offer the same VMs under many purchasing options that differ across their cost, performance, availability, and time commitments. Popular purchasing options include on-demand, reserved, and transient VM types. Reserved VMs require long time commitments, whereas users can acquire and release the on-demand (and transient) VMs at any time. While transient VMs cost significantly less than on-demand VMs, platforms may revoke them at any time. In general, the stronger …

Sensor Fusion For Object Detection And Tracking In Autonomous Vehicles, Mohamad Ramin Nabati

Doctoral Dissertations

Autonomous driving vehicles depend on their perception system to understand the environment and identify all static and dynamic obstacles surrounding the vehicle. The perception system in an autonomous vehicle uses the sensory data obtained from different sensor modalities to understand the environment and perform a variety of tasks such as object detection and object tracking. Combining the outputs of different sensors to obtain a more reliable and robust outcome is called sensor fusion. This dissertation studies the problem of sensor fusion for object detection and object tracking in autonomous driving vehicles and explores different approaches for utilizing deep neural networks …

Toward Reliable And Efficient Message Passing Software For Hpc Systems: Fault Tolerance And Vector Extension, Dong Zhong

Doctoral Dissertations

As the scale of High-performance Computing (HPC) systems continues to grow, researchers are devoted themselves to achieve the best performance of running long computing jobs on these systems. My research focus on reliability and efficiency study for HPC software.

First, as systems become larger, mean-time-to-failure (MTTF) of these HPC systems is negatively impacted and tends to decrease. Handling system failures becomes a prime challenge. My research aims to present a general design and implementation of an efficient runtime-level failure detection and propagation strategy targeting large-scale, dynamic systems that is able to detect both node and process failures. Using multiple overlapping …

Human Fatigue Predictions In Complex Aviation Crew Operational Impact Conditions, Suresh Rangan

Doctoral Dissertations

In this last decade, several regulatory frameworks across the world in all modes of transportation had brought fatigue and its risk management in operations to the forefront. Of all transportation modes air travel has been the safest means of transportation. Still as part of continuous improvement efforts, regulators are insisting the operators to adopt strong fatigue science and its foundational principles to reinforce safety risk assessment and management. Fatigue risk management is a data driven system that finds a realistic balance between safety and productivity in an organization. This work discusses the effects of mathematical modeling of fatigue and its …

Towards Secure Deep Neural Networks For Cyber-Physical Systems, Jiangnan Li

Doctoral Dissertations

In recent years, deep neural networks (DNNs) are increasingly investigated in the literature to be employed in cyber-physical systems (CPSs). DNNs own inherent advantages in complex pattern identifying and achieve state-of-the-art performances in many important CPS applications. However, DNN-based systems usually require large datasets for model training, which introduces new data management issues. Meanwhile, research in the computer vision domain demonstrated that the DNNs are highly vulnerable to adversarial examples. Therefore, the security risks of employing DNNs in CPSs applications are of concern.

In this dissertation, we study the security of employing DNNs in CPSs from both the data domain …

Utility Scale Building Energy Modeling And Climate Impacts, Brett C. Bass

Doctoral Dissertations

Energy consumption is steadily increasing year over year in the United States (US). Climate change and anthropogenically forced shifts in weather have a significant impact on energy use as well as the resilience of the built environment and the electric grid. With buildings accounting for about 40% of total energy use in the US, building energy modeling (BEM) at a large scale is critical. This work advances that effort in a number of ways. First, current BEM approaches, their ability to scale to large geographical areas, and global climate models are reviewed. Next, a methodology for large-scale BEM is illustrated, …

An Analysis Of Modern Password Manager Security And Usage On Desktop And Mobile Devices, Timothy Oesch

Doctoral Dissertations

Security experts recommend password managers to help users generate, store, and enter strong, unique passwords. Prior research confirms that managers do help users move towards these objectives, but it also identified usability and security issues that had the potential to leak user data or prevent users from making full use of their manager. In this dissertation, I set out to measure to what extent modern managers have addressed these security issues on both desktop and mobile environments. Additionally, I have interviewed individuals to understand their password management behavior.

I begin my analysis by conducting the first security evaluation of the …

Addressing Security Challenges In Embedded Systems And Multi-Tenant Fpgas, Georgios Provelengios

Doctoral Dissertations

Embedded systems and field-programmable gate arrays (FPGAs) have become crucial parts of the infrastructure that supports our modern technological world. Given the multitude of threats that are present, the need for secure computing systems is undeniably greater than ever. Embedded systems and FPGAs are governed by characteristics that create unique security challenges and vulnerabilities. Despite their array of uses, embedded systems are often built with modest microprocessors that do not support the conventional security solutions used by workstations, such as virus scanners. In the first part of this dissertation, a microprocessor defense mechanism that uses a hardware monitor to protect …