Engineering Commons^™

Reinforcement Learning Approach For Inspect/Correct Tasks, Hoda Nasereddin

LSU Doctoral Dissertations

In this research, we focus on the application of reinforcement learning (RL) in automated agent tasks involving considerable target variability (i.e., characterized by stochastic distributions); in particular, learning of inspect/correct tasks. Examples include automated identification & correction of rivet failures in airplane maintenance procedures, and automated cleaning of surgical instruments in a hospital sterilization processing department. The location of defects and the corrective action to be taken for each varies from task episode. What needs to be learned are optimal stochastic strategies rather than optimization of any one single defect type and location. RL has been widely applied in robotics …

Monitoring Of Remote Hydrocarbon Wells Using Azure Internet Of Things, Derek W. Staal

LSU Master's Theses

Remote monitoring of hydrocarbon wells is a tedious and meticulously thought out task performed to create a cyber-physical bridge between the asset and the owner. There are many systems and techniques on the market that offer this solution but due to their lack of interoperability and/or decentralized architecture they begin to fall apart when remote assets become farther away from the client. This results in extreme latency and thus poor decision making. Microsoft's Azure IoT Edge was the focus of this writing. Coupled with off-the-shelf hardware, Azure's IoT Edge services were integrated with an existing unit simulating a remote hydrocarbon …

An Approach To Counting Vehicles From Pre-Recorded Video Using Computer Algorithms, Mishuk Majumder

LSU Master's Theses

One of the fundamental sources of data for traffic analysis is vehicle counts, which can be conducted either by the traditional manual method or by automated means. Different agencies have guidelines for manual counting, but they are typically prepared for particular conditions. In the case of automated counting, different methods have been applied, but You Only Look Once (YOLO), a recently developed object detection model, presents new potential in automated vehicle counting. The first objective of this study was to formulate general guidelines for manual counting based on experience gained in the field. Another goal of this study was to …

Optimizing The Performance Of Multi-Threaded Linear Algebra Libraries Based On Task Granularity, Shahrzad Shirzad

LSU Doctoral Dissertations

Linear algebra libraries play a very important role in many HPC applications. As larger datasets are created everyday, it also becomes crucial for the multi-threaded linear algebra libraries to utilize the compute resources properly. Moving toward exascale computing, the current programming models would not be able to fully take advantage of the advances in memory hierarchies, computer architectures, and networks. Asynchronous Many-Task(AMT) Runtime systems would be the solution to help the developers to manage the available parallelism. In this Dissertation we propose an adaptive solution to improve the performance of a linear algebra library based on a set of compile-time …

Security Analysis Of The Anonabox Pro, Devi Sowjanya Vempati

LSU Master's Theses

The security posture of critical network appliances, such as routers, switches, and firewalls, is crucial as these devices provide the first line of defense against malicious actors. As there is typically no endpoint security software, such as anti-virus available for these devices, it is crucial that vendors release timely updates for vulnerabilities and that customers apply them quickly. The research effort documented in this thesis describes a developed methodology for triaging known vulnerabilities in network devices, assessing the security posture of device vendors, and locating likely weak points in released firmware images. To highlight this methodology, the Anonabox security appliance …

A Study On The Improvement Of Data Collection In Data Centers And Its Analysis On Deep Learning-Based Applications, Dipak Kumar Singh

LSU Doctoral Dissertations

Big data are usually stored in data center networks for processing and analysis through various cloud applications. Such applications are a collection of data-intensive jobs which often involve many parallel flows and are network bound in the distributed environment. The recent networking abstraction, coflow, for data parallel programming paradigm to express the communication requirements has opened new opportunities to network scheduling for such applications. Therefore, I propose coflow based network scheduling algorithm, Coflourish, to enhance the job completion time for such data-parallel applications, in the presence of the increased background traffic to mimic the cloud environment infrastructure. It outperforms …

Why Do You Take That Route?, Alimire Nabijiang

LSU Master's Theses

We conducted this study to investigate causal factors which affects the route-choice behavior of drivers. Due to the insufficiency of measured variables and the lack of high-fidelity properties obtained by conventional data, we carried out Stated Choice Experiment (SCE) using Immersive Virtual Environments (IVE) tool to collect data. IVE tool can provide realistic virtual experiences to participants, and give advantage of flexible environment for exploring causal factors of drivers’ decision on route choice. Based on reviewing literatures, we believe that there is limited studies that has tried to comprehend the route choice behavior of drivers using causal inference. The study …

Understanding And Optimizing Flash-Based Key-Value Systems In Data Centers, Yichen Jia

LSU Doctoral Dissertations

Flash-based key-value systems are widely deployed in today’s data centers for providing high-speed data processing services. These systems deploy flash-friendly data structures, such as slab and Log Structured Merge(LSM) tree, on flash-based Solid State Drives(SSDs) and provide efficient solutions in caching and storage scenarios. With the rapid evolution of data centers, there appear plenty of challenges and opportunities for future optimizations.

In this dissertation, we focus on understanding and optimizing flash-based key-value systems from the perspective of workloads, software, and hardware as data centers evolve. We first propose an on-line compression scheme, called SlimCache, considering the unique characteristics of key-value …

Retrospective Study Of Fmol Health System Utilization Using Geospatial Information, Deekshith Mandala

LSU Master's Theses

Medicaid Expansion and closing of Emergency Departments (ED) like Earl K. Long, Baton Rouge General Mid-City ED, and Champion Medical Center changed the health care landscape in East Baton Rouge Parish (EBRP). In this research study, a Geographical Information System (GIS) is used to analyze the impact of the expansion of Medicaid and the inauguration of Our Lady of the Lake North Baton Rouge ED (OLOL NBR ED) over the utilization of Franciscan Missionaries of our Lady Health System (FMOLHS) for both emergency and non-emergency health care services. This study is performed across the 58 neighborhoods of EBRP. Overutilization of …