Computer Engineering Commons^™

Research On Fire Emergency Evacuation Simulation Based On Cooperative Deep Reinforcement Learning, Lingjia Ni, Xiaoxia Huang, Hongga Li, Zibo Zhang

Journal of System Simulation

Abstract: The fire accident is a major threat to the public safety, in which the high temperature, toxic and harmful gases seriously interfer the selection of the evacuation routes. Deep reinforcement learning is introduced into the research of emergency evacuation simulation, and a cooperative double deep Q network algorithm is proposed for the multi-agent environment. A fire scene model that changes dynamically over time is established to provide the real-time information on the distribution of the dangerous areas for the evacuation. The independent agent neural networks are integrated and the multi-agent unified deep neural network is established to realize the …

Misinformation And Disinformation: Detecting Fakes With The Eye And Ai, Victoria Rubin

Data and Test Instruments

How do we detect, deter, and prevent the spread of mis- and disinformationwith the human eye and AI? How does theory inform the practice, and how do theevidence-based research and best practices in lie-catching and truth-seekingprofessions—inform AI? The book looks into well-established human practicessuch as the routines and processes used in detective work, journalism, and scientificinquiry, and how they contribute toward innovative AI solutions. The book explainsthe principles, inner workings, and recent evolution of five types of state-of-the-artAI technologies suitable for curtailing the spread of mis- and disinformation:automated deception detectors, clickbait detectors, satirical fake detectors, rumordebunkers, and computational fact-checking tools.

Machine Learning, The Mortar Of Modernization, Vishnu Pendyala

Open Educational Resources

Machine Learning is advancing civilization and is one of the key drivers of the economy today. Machine Learning literacy may one day become essential, the way computer literacy is today. This talk intends to explain Machine Learning concepts to wider audiences by using easy-to-relate real-world analogies and serves as a refresher to those already initiated. A fundamental concept in machine learning is similarity. The dot product that is ubiquitously present in machine learning is a measure of similarity. Similarity is key to human learning as well. We learn in delta increments by comparing and contrasting with what we already know. …

Container Migration In Ad-Hoc Wireless Mesh Networks, Gagan Gupta, Kyle Fenole, Siddharth Venkatesh

Computer Science and Engineering Senior Theses

This project aims to implement container migration on an ad hoc mesh network, which would allow for a mobile fog computing mesh network in areas with no connectivity. This kind of system would allow for a rapidly deployable and mobile compute cluster that could be used in rural scenarios or situations where existing connectivity infrastructures are out of commission i.e. disaster recovery. We use technologies like Docker, CRIU, and batman-adv to create a mesh compute network that is decentralized, flexible, and consistent. Docker with CRIU facilitates container live migration by checkpointing a container in one network node and restoring that …

Velachain: A Decentralized Exchange Built For Cross-Chain Communication, Connor Callahan, Bradley Lostak

Computer Science and Engineering Senior Theses

The blockchain industry is one of the fastest growing industries in the world right now. Billions of dollars of venture capital money is being put to work to innovate in this new space. One of the problems that has plagued the blockchain space since its inception is the lack of interoperability between blockchains. A single blockchain is great at dictating its own state but they rarely contain functionality to communicate with other blockchains. Furthermore, users on these blockchains have limited ways to swap cryptocurrencies from one blockchain to another. For this project, we built a cryptocurrency decentralized exchange built on …

Method And System For Automatic Selection Of One Or More Image Processing Algorithm, Tanushyam Chattopadhyay, Ramu Vempada Reddy, Utpal Garain

Patents

Disclosed is a method and system for automatic algorithm selection for image processing. The invention discloses the method and system for automatically selecting the correct algorithm(s) for a varying requirement of the image for processing. The selection of algorithm is completely automatic and guided by a plurality of machine learning approaches. The system here is configured to pre-process plurality of images for creating a training data. Next, the test image is extracted, pre-processed and matched for assessing the best possible match of algorithm for processing.

Brilliance Bias In Gpt-3, Ashley Troske, Edith Gonzalez, Nicole Lawson

Computer Science and Engineering Senior Theses

Language has a profound impact on how we perceive the world. With GPT- 3’s rise in popularity, present in 300 applications averaging 4.5 billion words per day, it is critical for us as programmers to identify and correct biases in its generations. A variety of biases have been identified in generative language models, spanning biases based on gender, race, and religion. Our project pioneers the study of the Brilliance Bias for generative models. This implicit, yet powerful bias imposes the idea of “brilliance” being a male trait and in turn, sets back women’s achievements starting as young as 5-7 years. …

Database Of Non-Government Organizations On The Global Scale, Jeremy Mekker

Computer Science and Engineering Senior Theses

Currently in the world, there are about 10 million non-profit/non-governmental organizations. In order to find these organizations, one must diligently scour the Internet for hours on end. Most research into helping refugees or those in need results in only finding larger organizations that cover many different tasks in the targeted area. This is an active problem. While larger organizations tend to help a wider audience of people, often the help that one would like to give does not get delivered as urgently as the donor would hope. In order to combat this, the primary objective would be to optimize the …

Neural Network Interpretability For Autonomous Driving Neural Networks, Raghav Kapoor, Casey Nguyen

Computer Science and Engineering Senior Theses

In the field of neural networks, there has been a long-standing problem that needs to be addressed: gaining insight into how neural networks make decisions. Neural Networks are still considered black boxes and are often difficult to understand. This lack of understanding becomes an ethical dilemma especially in the domain of self-driving cars. Given the limited number of works geared towards unravelling neural network logic for autonomous driving vehicles, our team seeks to create a novel neural network interpretability method to influence the neural network during its training process.

Deep Neural Networks have demonstrated impressive performance in complex tasks, such …

Improving Diversity In Journalistic Sources With Computer Vision, Austin Johnson, Carlos Mercado, Sabiq Khan

Computer Science and Engineering Senior Theses

News is a constant part of our lives, and has a significant impact on how we see the world. Simply put, journalistic sources are not diverse enough, and typically are only those in the majority. Newsrooms have made passionate declarations about wanting more diversity in their news sources. Our project aims to help with that. Building on top of the DEI toolkit, we aim to help newsrooms increase the diversity of their sources by analyzing their sources using a google search to pick an image and facial recognition software to determine the gender and race of the image.

Mysyllabi: An Online Syllabus Repository, Aristos Xanthus

Computer Science and Engineering Senior Theses

College students need more information when selecting coursework. During the time of course registration, students are limited to the course name, time, description, and professor. To make the best course selection, students need access to the syllabus which provides a comprehensive overview of the course as well as course-specific details such as the structure, workload, and teaching style. However, syllabi are hard to obtain before the first day of class, and they are seldom found online. Therefore, I propose MySyllabi, an online syllabus repository as a quick, accessible, and reliable location to find syllabi for a university’s classes/professors that will …

Expanding Neuro-Symbolic Artificial Intelligence For Strategic Learning, Dorian Clay

Computer Science and Engineering Senior Theses

Today, artificial intelligence is used to solve problems as varied as driving cars to diagnosing diseases. However, it still has major pitfalls. The most common form of AI today, machine learning, can produce near-certain predictions for some tasks, yet completely fail at a related but different task. Additionally, such algorithms cannot do reasoning for complex problems, are not well explainable, require massive training datasets, and are sensitive to new data.

One solution to these problems is combining today’s cutting-edge neural networks with an older idea in AI: symbolic AI. Such so-called “neuro-symbolic AI” systems can be understood at a high …

Netcon: Dynamic Resource Allocation To Containers Running On Network Switching Appliances, Siena Hanna, Justin Lee, Tania Pham

Computer Science and Engineering Senior Theses

A WiFi Access Point (AP) is an important technology that switches data packets to transmit and receive wireless signals to create WiFi. These Access Points (AP) can have powerful processors, but they are idle most of the time [1]. Instead, these idle processing resources could be redirected to run containers for low-latency applications. However, it is essential to ensure that the AP does not suffer performance issues as a result of running containers. The main function of APs is to switch packets, and this function must be preserved while containers are running. In this thesis, we propose a method, NetCon, …

Smartnic Compatible Blockchain, Gavin Jampani, Eish Kapoor, Vivek Ponnala

Computer Science and Engineering Senior Theses

The intent of our Senior Design Thesis was to develop a software infrastructure for Blockchain applications that is compatible with SmartNIC (smart network interface card) hardware to contribute to the future of edge computing and Blockchain. The demand on network infrastructure is exploding, with there estimated to be 41.2 billion active network connections in 2025, 30.8 of those being IOT [13]. Of these active network IOT connections, a growing plurality will contain SmartNICs, which add valuable compute into the crucial intersection between the network and individual server. With it forecasted that three-fourths of all enterprise data will be processed at …

Enhancing The Quality Of Service And Energy Efficiency Of Wifi-Based Iot Networks, Jaykumar Sheth

Engineering Ph.D. Theses

The 802.11 standard, known as WiFi, is currently being used for a wide variety of applications including Internet of Things (IoT). However, the contention between the traffic of IoT stations (STAs) as well as the contention between these flows and regular user-generated traffic reduces the energy efficiency and timeliness of IoT communication. To remedy this problem, in this thesis, we take the following approaches for mitigating the challenges faced by WiFi-based IoT networks: First, we highlight the importance of observability with respect to WiFi networks and how it helps the researchers to better examine the dynamics of issues and its …

Sengasafe, Patrick Zhang, Matthew Hall, Matthew Brunkhorst

Computer Science and Engineering Senior Theses

There is a lack of effective education about sexual health for women in rural Uganda. Compared to the rest of the world, Uganda has a very high incidence rate of cervical cancer. This is much higher than it should be because it is well established that the human papillomavirus (HPV) is the primary cause of cervical cancer, and vaccinations for HPV have been available in rural Uganda since 2006. However, most rural Ugandan women are not vaccinated due to poor vaccination delivery, negative attitudes towards vaccination, and other social cultural factors. The existing sexual education system run by the Ugandan …

Donation Pick-Up And Tracking App For Columbian Food Bank, Alex Fang, Arren Leung, Adrian Ramirez Lopez, Rodrigo Mejia

Computer Science and Engineering Senior Theses

Banco de Alimentos Cali is responsible for picking up food donations from local businesses, organizing by perishable dates, and distributing them to the communities in need. When donations are picked up from the donor, the bank is responsible for sending their donors a certificate with the donation's information, including the dollar equivalent of their donation. This certificate is sent out only if the donor requests it for tax deduction purposes. While this plan may seem simple in theory, the food bank often struggles to get the necessary information from their drivers in regards to the donation, which ultimately delays the …

Halide Perovskites Enable Polaritonic Xy Spin Hamiltonian At Room Temperature, Renjie Tao, Kai Peng, Louis Haeberlé, Quanwei Li, Dafei Jin, Graham R. Fleming, Stéphane Kéna-Cohen, Xiang Zhang, Wei Bao

Department of Electrical and Computer Engineering: Faculty Publications

Exciton polaritons, the part-light and part-matter quasiparticles in semiconductor optical cavities, are promising for exploring Bose–Einstein condensation, non-equilibrium many-body physics and analogue simulation at elevated temperatures. However, a room-temperature polaritonic platform on par with the GaAs quantum wells grown by molecular beam epitaxy at low temperatures remains elusive. The operation of such a platform calls for long-lifetime, strongly interacting excitons in a stringent material system with large yet nanoscale-thin geometry and homogeneous properties. Here, we address this challenge by adopting a method based on the solution synthesis of excitonic halide perovskites grown under nanoconfinement. Such nanoconfinement growth facilitates the synthesis …

Generating Reality-Analogous Datasets For Autonomous Uav Navigation Using Digital Twin Areas, Thomas Lee, Susan Mckeever, Jane Courtney

Conference papers

In order for autonomously navigating Unmanned Air Vehicles(UAVs) to be implemented in day-to-day life, proof of safe operation will be necessary for all realistic navigation scenarios. For Deep Learning powered navigation protocols, this requirement is challenging to fulfil as the performance of a network is impacted by how much the test case deviates from data that the network was trained on. Though networks can generalise to manage multiple scenarios in the same task, they require additional data representing those cases which can be costly to gather. In this work, a solution to this data acquisition problem is suggested by way …

Easycompost: Using Augmented Reality To Improve Reality, Jonathan Nguyen, Andrew Bredar

Computer Science and Engineering Senior Theses

In the United States, a significant portion of landfill is undiverted compost and recyclable material. Instead of being sent to landfills, compostable material can be used as nutrient rich fertilizer and soil in local communities. Likewise, recyclable material can be processed at appropriate facilities to reuse materials and reduce the demand for unsustainable resources, like plastics. We seek to reduce the portion of undiverted recyclable and compostable waste by aiding in the proper sorting of waste at common sites of disposal, particularly trash cans in cafeterias and restaurants. We have designed a mobile, augmented reality, 3D object detection model that …

Wage Wizard, Jack Davey, Kyle Felip Mondina, Brett Rimmer

Interdisciplinary Design Senior Theses

Wage theft is a severe problem in Santa Clara County, with the Santa Clara Wage Theft Coalition identifying more than 25,000 local wage theft cases as of 20211. In this study we will be addressing the issue of wage theft particularly in the Santa Clara County caregiving industry. Wage theft is prevalent in the caregiving industry for several reasons. First, the industry employs many migrant Filipino workers who are unaware of their legal rights and protections or even the fact if they are experiencing wage theft or not. Second, the sizable undocumented portion of these workers are coerced to lower …

Orient: Teaching Object Oriented Programming With Augmented Reality, Shivangi Kar, Stephanie Lu

Computer Science and Engineering Senior Theses

In our technologically advanced society, computational thinking is a critical skill for students to develop. Our project, ORIENT, is a mobile application that uses augmented reality to teach object-oriented programming(OOP), a fundamental concept in computer science. ORIENT is designed for novice programmers from the middle school level up, and it consists of a three-part tutorial series that teaches class creation, inheritance, and polymorphism-three of the most important tenets of object-oriented programming. ORIENT, which was built in Unity and made specifically for the iPad, includes a series of interactive tutorials with immediate feedback; it encourages students to explore OOP in a …

Brass: Bandwidth Reservation On Arista Switching Systems, Angelus Mcnally, Conrad Park, Tyler Tivadar

Computer Science and Engineering Senior Theses

As networking technologies in the fields of Cloud, Fog, and Edge computing continue to gain popularity and widespread adoption, their contributions to network traffic load and congestion in their respective data processing layers are expected to increase. The implication of this inevitable change is that overall traffic congestion is expected to increase as more devices are connected. As a consequence of this congestion, network performance is expected to decrease, and the best effort nature of network traffic means that there is no guarantee that applications will continue to communicate at rates required for operation. This could prevent certain networking applications …

Method And System For Region Proposal Based Object Recognition For Estimating Planogram Compliance, Avishek Kumar Shaw, Rajashree Ramakrishnan, Shilpa Yadukumar Rao, Pranoy Hari, Dipti Prasad Mukherjee, Bikash Santra

Patents

No abstract provided.

Deep Learning Pseudocode Generation: A Qualitative Analysis, Colin Rioux

Computer Science and Engineering Master's Theses

Pseudocode is a traditional teaching tactic in computer science, yet it is not standardized and programming language dependent. Thus, it can be quite time consuming to write it. With the advancement of AI methodologies in NLP, AI could help address this problem. This work investigates the quality of AI generated pseudocode from source code. Five studies are conducted in this work to measure pseudocode quality, where each study modifies model input to observe accuracy and generalizability. The results show that there is an association between pseudocode quality and training and test set similarity. Furthermore, a sizable and diverse training set …

The Future Of Camera Technology With Atul Ingle, Atul Ingle

PDXPLORES Podcast

Single-photon camera sensors have the potential to revolutionize digital camera technology. These sensors can capture individual photons at high speeds, circumventing some of the limitations of current digital camera technologies. One challenge, however, is the sheer amount of data generated by the sensor, which is a hindrance to their widespread adoption. With the support of the National Science Foundation, Assistant Professor Atul Ingle is developing algorithms to solve this problem--algorithms that could enable the adoption of single-photon sensors in applications ranging from autonomous vehicles to medical imaging and the cameras in cellphones.

Knowledge-Driven Drug-Use Namedentity Recognition With Distant Supervision, Goonmeet Bajaj, Ugur Kursuncu, Manas Gaur, Usha Lokala, Ayaz Hyder, Srinivasan Parthasarathy, Amit Sheth

Publications

As Named Entity Recognition (NER) has been essential in identifying critical elements of unstructured content, generic NER tools remain limited in recognizing entities specific to a domain, such as drug use and public health. For such high-impact areas, accurately capturing relevant entities at a more granular level is critical, as this information influences real-world processes. On the other hand, training NER models for a specific domain without handcrafted features requires an extensive amount of labeled data, which is expensive in human effort and time. In this study, we employ distant supervision utilizing a domain-specific ontology to reduce the need for …

Emt: Software-Based Energy Management Tool For Wifi Iot Devices, Aastha Chawla, Sreya Goyalia, Nidusha Kannan

Computer Science and Engineering Senior Theses

Many WiFi-based IoT (Internet of Things) devices rely on limited energy resources such as batteries or energy harvesting. Although monitoring and studying the energy consumption of these devices is essential, the use of external, hardware-based energy measurement tools is costly, non-scalable, and introduces many challenges regarding the connectivity of such tools with devices. In this thesis, we propose EMT, a novel tool to collect, analyze, and monitor the power cycles of IoT devices without the need for any external tools. The basic idea is to modify the WiFi Access Point's software to keep track of the power status of devices …

Unstumm | Augmented Voyage: A Platform For Telematic Live Performances In Augmented Reality, Unstumm Unstumm, Nicola Hein, Claudia Schmitz, Sven Hahne Hahne

Journal of Network Music and Arts

UnStumm | Augmented Voyage is curated and developed by time-based media artist Claudia Schmitz and sound artist and guitarist Nicola L. Hein. It was programmed and developed by software engineer Sven Hahne. UnStumm | Augmented Voyage is also a mobile app and server infrastructure that serves as the artistic vehicle to realize telematic live performances with artists from around the globe. It consists of video art, music, and dance in augmented reality.
Through the use of our software, the audience is able to view the video streams by video artists as 3D video sculptures and listen to the audio streams …

“Dancedemic” In A Pandemic: A New Networked Reality, Ellen Pearlman

Journal of Network Music and Arts

The Covid-19 pandemic of 2020 forced large parts of the globe and entire industries to shut down. Network connectivity became a necessary lifeline deploying a host of technologies that, while once considered experimental, were to be adopted as a vital part of just-in-time solutions-based practices. This report examines the implementation of one of these topologies in the form of an artist-built, artist-owned global network in the hopes that other small-sized organizations can benefit from these ad hoc, but successful solutions.