Artificial Intelligence and Robotics Commons^™

Feature Selection From Clinical Surveys Using Semantic Textual Similarity, Benjamin Warner

McKelvey School of Engineering Graduate Student Theses & Dissertations

Survey data collected from human subjects can contain a high number of features while having a comparatively low quantity of examples. Machine learning models that attempt to predict outcomes from survey data under these conditions can overfit and result in poor generalizability. One remedy to this issue is feature selection, which attempts to select an optimal subset of features to learn upon. A relatively unexplored source of information in the feature selection process is the usage of textual names of features, which may be semantically indicative of which features are relevant to a target outcome. The relationships between feature names …

An Empirical Study Of Pre-Trained Model Reuse In The Hugging Face Deep Learning Model Registry, Wenxin Jiang, Nicholas Synovic, Matt Hyattt, Taylor R. Schorlemmer, Rohan Sethi, Yung-Hisang Lu, George K. Thiruvathukal, James C. Davis

Computer Science: Faculty Publications and Other Works

Deep Neural Networks (DNNs) are being adopted as components in software systems. Creating and specializing DNNs from scratch has grown increasingly difficult as state-of-the-art architectures grow more complex. Following the path of traditional software engineering, machine learning engineers have begun to reuse large-scale pre-trained models (PTMs) and fine-tune these models for downstream tasks. Prior works have studied reuse practices for traditional software packages to guide software engineers towards better package maintenance and dependency management. We lack a similar foundation of knowledge to guide behaviors in pre-trained model ecosystems.

In this work, we present the first empirical investigation of PTM reuse. …

Optimizing Tumor Xenograft Experiments Using Bayesian Linear And Nonlinear Mixed Modelling And Reinforcement Learning, Mary Lena Bleile

Statistical Science Theses and Dissertations

Tumor xenograft experiments are a popular tool of cancer biology research. In a typical such experiment, one implants a set of animals with an aliquot of the human tumor of interest, applies various treatments of interest, and observes the subsequent response. Efficient analysis of the data from these experiments is therefore of utmost importance. This dissertation proposes three methods for optimizing cancer treatment and data analysis in the tumor xenograft context. The first of these is applicable to tumor xenograft experiments in general, and the second two seek to optimize the combination of radiotherapy with immunotherapy in the tumor xenograft …

Identifying Key Activity Indicators In Rats' Neuronal Data Using Lasso Regularized Logistic Regression, Avery Woods

Honors Theses

This thesis aims to identify timestamps of rats’ neuronal activity that best determine behavior using a machine learning model. Neuronal data is a complex and high-dimensional dataset, and identifying the most informative features is crucial for understanding the underlying neuronal processes. The Lasso regularization technique is employed to select the most relevant features of the data to the model’s prediction. The results of this study provide insights into the key activity indicators that are associated with specific behaviors or cognitive processes in rats, as well as the effect that stress can have on neuronal activity and behavior. Ultimately, it was …

Automatic Identification Of Jetting Behavior In 3d Printing With Binary Classification And Anomaly Detection, Alexander Chandy

Honors Scholar Theses

Consistently jetting different materials from the print head of a 3D printer is a key, yet challenging task in manufacturing processes. By using active machine learning, we can efficiently predict complex diagrams that illustrate the region of printing conditions under which “desirable jetting”, “jetting”, and “no jetting” of ink occurs for different substances. However, labeling the images of printed ink droplets that are fed to the active learning model can be time intensive. Therefore, it is ideal to use computer vision to automate the classification of this image data. This classification can be broken down into two steps. In the …

U-No: U-Shaped Neural Operators, Md Ashiqur Rahman, Zachary E Ross, Kamyar Azizzadenesheli

Department of Computer Science Faculty Publications

Neural operators generalize classical neural networks to maps between infinite-dimensional spaces, e.g., function spaces. Prior works on neural operators proposed a series of novel methods to learn such maps and demonstrated unprecedented success in learning solution operators of partial differential equations. Due to their close proximity to fully connected architectures, these models mainly suffer from high memory usage and are generally limited to shallow deep learning models. In this paper, we propose U-shaped Neural Operator (U-NO), a U-shaped memory enhanced architecture that allows for deeper neural operators. U-NOs exploit the problem structures in function predictions and demonstrate fast training, data …

Pruning Ghsom To Create An Explainable Intrusion Detection System, Thomas Michael Kirby

Theses and Dissertations

Intrusion Detection Systems (IDS) that provide high detection rates but are black boxes lead
to models that make predictions a security analyst cannot understand. Self-Organizing Maps
(SOMs) have been used to predict intrusion to a network, while also explaining predictions through
visualization and identifying significant features. However, they have not been able to compete with
the detection rates of black box models. Growing Hierarchical Self-Organizing Maps (GHSOMs)
have been used to obtain high detection rates on the NSL-KDD and CIC-IDS-2017 network traffic
datasets, but they neglect creating explanations or visualizations, which results in another black
box model.
This paper offers …

Development And Evaluation Of An Automated Tactical Tillage Tool To Control Weeds In Row-Crop Production Systems, Grace Mccormick Friday

Theses and Dissertations

Weed control is an integral part of a successful overall production strategy in row- cropping systems and has the potential to reduce or eliminate yield losses that negatively affect profitability. Timely and correctly selected herbicide applications are the major keys for effective weed control in a majority of instances. However, there are negative factors that contribute to ineffectiveness and weed escape issues that currently lack viable options for management. Sparsely populated late-season weeds that emerge after lay-by herbicide applications and weeds that have become tolerant and resistant to traditional herbicide chemistries are of greatest concern. Historically, these weeds would have …

Secure And Efficient Federated Learning, Xingyu Li

Theses and Dissertations

In the past 10 years, the growth of machine learning technology has been significant, largely due to the availability of large datasets for training. However, gathering a sufficient amount of data on a central server can be challenging. Additionally, with the rise of mobile networking and the large amounts of data generated by IoT devices, privacy and security issues have become a concern, resulting in government regulations such as GDPR, HIPAA, CCPA, and ADPPA. Under these circumstances, traditional centralized machine learning methods face a problem in that sensitive data must be kept locally for privacy reasons, making it difficult to …

Tornado Outbreak False Alarm Probabilistic Forecasts With Machine Learning, Kirsten Reed Snodgrass

Theses and Dissertations

Tornadic outbreaks occur annually, causing fatalities and millions of dollars in damage. By improving forecasts, the public can be better equipped to act prior to an event. False alarms (FAs) can hinder the public’s ability (or willingness) to act. As such, a probabilistic FA forecasting scheme would be beneficial to improving public response to outbreaks.

Here, a machine learning approach is employed to predict FA likelihood from Storm Prediction Center (SPC) tornado outbreak forecasts. A database of hit and FA outbreak forecasts spanning 2010 – 2020 was developed using historical SPC convective outlooks and the SPC Storm Reports database. Weather …

Transformer-Based Feature Fusion Approach For Multimodal Visual Sentiment Recognition Using Tweets In The Wild, Fatimah Alzamzami, Abdulmotaleb El Saddik

Computer Vision Faculty Publications

We present an image-based real-time sentiment analysis system that can be used to recognize in-the-wild sentiment expressions on online social networks. The system deploys the newly proposed transformer architecture on online social networks (OSN) big data to extract emotion and sentiment features using three types of images: images containing faces, images containing text, and images containing no faces/text. We build three separate models, one for each type of image, and then fuse all the models to learn the online sentiment behavior. Our proposed methodology combines a supervised two-stage training approach and threshold-moving method, which is crucial for the data imbalance …

Integrating Ai Into Culinary Medicine: A Revolution In Nutrition And Home Cooking, Emeka Ikeakanam, Evan Curry, Terrence Mchugh, Jason Walker

Research Day

Introduction

With the growing popularity of the emerging field of culinary medicine, there is a growing understanding of the culinary barriers needed to be overcome to adopt healthier eating habits. Lack of confidence, low skills, and lack of time are some of the most common barriers that prevent individuals from cooking at home. However, integrating AI can offer personalized support for home cooking and help individuals overcome these barriers. AI-powered meal planning and recipe suggestions can guide healthy and nutritious food choices that cater to their dietary needs and preferences. Additionally, AI can modify recipes to accommodate individual health conditions …

Product Review Classification Using Machine Learning And Statistical Data Analysis, Kajal Singh

Independent Student Projects and Publications

The aim of the paper is to implement and analyze the machine learning models for product review dataset. The project focuses on binary classification, multi-class classification, and clustering approaches to analyze and categorize product reviews. The performance of the models over each of the five classification tasks is measured by the 5-fold cross-validation scores over the training data.

Soft Law 2.0: An Agile And Effective Governance Approach For Artificial Intelligence, Gary E. Marchant, Carlos Ignacio Gutierrez

Minnesota Journal of Law, Science & Technology

No abstract provided.

Digital Dna: The Ethical Implications Of Big Data As The World’S New-Age Commodity, Clark H. Dotson

Honors Theses

In the emerging digital world that we find ourselves in, it becomes apparent that data collection has become a staple of daily life, whether we like it or not. This research discussion aims to bring light to just how much one’s own digital identity is valued in the technologically-infused world of today, with distinct research and local examples to bring awareness to the ethical implications of your online presence. The paper in question examines anecdotal and research evidence of the collection of data, both through true and unjust means, as well as ethical implications of what this information truly represents. …

Programming An Autonomous Robot, Maxwell Brueggeman

Honors College Theses

Ravaged by hurricanes, Florida needed help restoring its natural beauty and returning its wildlife to their homes. This was the task for the IEEE SoutheastCon 2023 Hardware Competition. Florida’s restoration was simulated by returning various ducks and pillars that lay strewn across a game board to their proper places. Ducks needed to return to their pond, pillars needed to be stacked to create statues, and food needed to be placed in the manatee and alligator aquariums. Competing teams were challenged to create an autonomous robot capable of performing these tasks. During the first semester, sensor selection was tackled. Research was …

Fair Enough: Standardizing Evaluation And Model Selection For Fairness Research In Nlp, Xudong Han, Timothy Baldwin, Trevor Cohn

Natural Language Processing Faculty Publications

Modern NLP systems exhibit a range of biases, which a growing literature on model debiasing attempts to correct. However, current progress is hampered by a plurality of definitions of bias, means of quantification, and oftentimes vague relation between debiasing algorithms and theoretical measures of bias. This paper seeks to clarify the current situation and plot a course for meaningful progress in fair learning, with two key contributions: (1) making clear inter-relations among the current gamut of methods, and their relation to fairness theory; and (2) addressing the practical problem of model selection, which involves a trade-off between fairness and accuracy …

Analysis Of A Federated Learning Framework For Heterogeneous Medical Image Data: Privacy And Performance Perspective, Julia Brixey

Computer Science and Computer Engineering Undergraduate Honors Theses

The massive amount of data available in our modern world and the increase of computational efficiency and power have allowed for great advancements in several fields such as computer vision, image processing, and natural languages. At the center of these advancements lies a data-centric learning approach termed deep learning. However, in the medical field, the application of deep learning comes with many challenges. Some of the fundamental challenges are the lack of massive training datasets, unbalanced and heterogenous data between health applications and health centers, security and privacy concerns, and the high cost of wrong inference and prediction. One of …

Characterization Of 2d Quantum Materials Using Ai And Large-Scale Quantum Data Collection, Apoorva Bisht

Computer Science and Computer Engineering Undergraduate Honors Theses

2D materials like hexagonal boron nitride, graphene, and tungsten diselenide are widely utilized for studying their unique mechanical and opto-electronic properties to exploit them to make transistors and fabricating a variety of other devices. All these applications require that the 2D materials used be of specific uniform thickness. Until very recently, this process has been largely manual and tedious. However, few applications exploit the characteristic color-to-thickness correspondence of these near-transparent materials. To continue this effort, in this work we create a large-scale dataset for three different materials (hBN, graphene, and WSe$_2$) to train and test an image segmentation model along …

Physics-Based Human-In-The-Loop Machine Learning Combined With Genetic Algorithm Search For Multi-Criteria Optimization: Electrochemical Co2 Reduction Reaction, Naohiro Fujinuma, Samuel Lofland

College of Science & Mathematics Departmental Research

Machine learning (ML) can be a powerful tool to expedite materials research, but the deployment for experimental research is often hindered by data scarcity and model uncertainty. An human-in-the-loop procedure to tailor the implementation of ML for multicriteria optimization is described. The effectiveness of this procedure in the development of a nafion-based membrane electrode assembly for electrochemical CO2 reduction reaction (CO2RR) into CO for two targets is demonstrated: energy efficiency (EE) and partial current density for CO2RR (). Model-agnostic nonlinear correlation analyses identify the 11 features relevant to those targets. The three studied decision tree-based ML models yield similar cross-validation …

Improving The Performance Of Automatic Target Recognition Using Data Augmentation., Rolando Holmes

Electronic Theses and Dissertations

Automatic Target Recognition (ATR) is a task that aims to recognize targets based on data obtained from visual sensors such as Infrared cameras. Machine learning, in particular Deep Learning, has been the dominate approach to solve this task. However, in order to achieve high performance, Deep Learning requires a large amount of data that may not be available, which poses an issue when developing robust models. In this thesis, we propose the use of data augmentation from an automatic detector as an alternative source to supplement the need for more training data. We use a two stage approach to develop …

Modeling, Simulation And Control Of Microrobots For The Microfactory., Zhong Yang

Electronic Theses and Dissertations

Future assembly technologies will involve higher levels of automation in order to satisfy increased microscale or nanoscale precision requirements. Traditionally, assembly using a top-down robotic approach has been well-studied and applied to the microelectronics and MEMS industries, but less so in nanotechnology. With the boom of nanotechnology since the 1990s, newly designed products with new materials, coatings, and nanoparticles are gradually entering everyone’s lives, while the industry has grown into a billion-dollar volume worldwide. Traditionally, nanotechnology products are assembled using bottom-up methods, such as self-assembly, rather than top-down robotic assembly. This is due to considerations of volume handling of large …

The Six Emotional Dimension (6de) Model: A Multidimensional Approach To Analyzing Human Emotions And Unlocking The Potential Of Emotionally Intelligent Artificial Intelligence (Ai) Via Large Language Models (Llm), Jay Ratican, James Hutson

Faculty Scholarship

The rapid advancements in artificial intelligence (AI) research, particularly in training large language models (LLMs) such as OpenAI's ChatGPT 3.5 and 4, hold significant potential for future applications in education, healthcare, and assisted living. Emotionally intelligent AI systems can provide personalized and adaptive educational experiences, enhancing engagement and educational outcomes. In healthcare, they can offer empathetic mental health support, augmenting existing resources. In assisted living, AI companions can provide emotional support, cognitive stimulation, and monitoring services, promoting independence and safety. However, ethical considerations and privacy safeguards are crucial to ensure responsible deployment. Integrating emotionally intelligent AI in these domains has …

Contrabert: Enhancing Code Pre-Trained Models Via Contrastive Learning, Shangqing Liu, Bozhi Wu, Xiaofei Xie, Guozhu Meng, Yang. Liu

Research Collection School Of Computing and Information Systems

Large-scale pre-trained models such as CodeBERT, GraphCodeBERT have earned widespread attention from both academia and industry. Attributed to the superior ability in code representation, they have been further applied in multiple downstream tasks such as clone detection, code search and code translation. However, it is also observed that these state-of-the-art pre-trained models are susceptible to adversarial attacks. The performance of these pre-trained models drops significantly with simple perturbations such as renaming variable names. This weakness may be inherited by their downstream models and thereby amplified at an unprecedented scale. To this end, we propose an approach namely ContraBERT that aims …

Neural Episodic Control With State Abstraction, Zhuo Li, Derui Zhu, Yujing Hu, Xiaofei Xie, Lei Ma, Yan Zheng, Yan Song, Yingfeng Chen, Jianjun Zhao

Research Collection School Of Computing and Information Systems

Existing Deep Reinforcement Learning (DRL) algorithms suffer from sample inefficiency.Generally, episodic control-based approaches are solutions that leveragehighly-rewarded past experiences to improve sample efficiency of DRL algorithms.However, previous episodic control-based approaches fail to utilize the latentinformation from the historical behaviors (e.g., state transitions, topological similarities,etc.) and lack scalability during DRL training. This work introducesNeural Episodic Control with State Abstraction (NECSA), a simple but effectivestate abstraction-based episodic control containing a more comprehensive episodicmemory, a novel state evaluation, and a multi-step state analysis. We evaluate ourapproach to the MuJoCo and Atari tasks in OpenAI gym domains. The experimentalresults indicate that NECSA achieves higher …

Wearing Masks Implies Refuting Trump?: Towards Target-Specific User Stance Prediction Across Events In Covid-19 And Us Election 2020, Hong Zhang, Haewoon Kwak, Wei Gao, Jisun An

Research Collection School Of Computing and Information Systems

People who share similar opinions towards controversial topics could form an echo chamber and may share similar political views toward other topics as well. The existence of such connections, which we call connected behavior, gives researchers a unique opportunity to predict how one would behave for a future event given their past behaviors. In this work, we propose a framework to conduct connected behavior analysis. Neural stance detection models are trained on Twitter data collected on three seemingly independent topics, i.e., wearing a mask, racial equality, and Trump, to detect people’s stance, which we consider as their online behavior in …

Generative Stresnet For Crime Prediction, Ba Phong Tran, Hoong Chuin Lau

Research Collection School Of Computing and Information Systems

In this work, we combine STResnet (Zhang et al., 2017) with VAE Kingma & Welling (2013) to generate crime distribution. The outputs can be used for downstream tasks such as patrol deployment planning Chase et al. (2021).

Techsumbot: A Stack Overflow Answer Summarization Tool For Technical Query, Chengran Yang, Bowen Xu, Jiakun Liu, David Lo

Research Collection School Of Computing and Information Systems

Stack Overflow is a popular platform for developers to seek solutions to programming-related problems. However, prior studies identified that developers may suffer from the redundant, useless, and incomplete information retrieved by the Stack Overflow search engine. To help developers better utilize the Stack Overflow knowledge, researchers proposed tools to summarize answers to a Stack Overflow question. However, existing tools use hand-craft features to assess the usefulness of each answer sentence and fail to remove semantically redundant information in the result. Besides, existing tools only focus on a certain programming language and cannot retrieve up-to-date new posted knowledge from Stack Overflow. …

Fine-Grained Commit-Level Vulnerability Type Prediction By Cwe Tree Structure, Shengyi Pan, Lingfeng Bao, Xin Xia, David Lo, Shanping Li

Research Collection School Of Computing and Information Systems

Identifying security patches via code commits to allow early warnings and timely fixes for Open Source Software (OSS) has received increasing attention. However, the existing detection methods can only identify the presence of a patch (i.e., a binary classification) but fail to pinpoint the vulnerability type. In this work, we take the first step to categorize the security patches into fine-grained vulnerability types. Specifically, we use the Common Weakness Enumeration (CWE) as the label and perform fine-grained classification using categories at the third level of the CWE tree. We first formulate the task as a Hierarchical Multi-label Classification (HMC) problem, …

Chronos: Time-Aware Zero-Shot Identification Of Libraries From Vulnerability Reports, Yunbo Lyu, Thanh Le Cong, Hong Jin Kang, Ratnadira Widyasari, Zhipeng Zhao, Xuan-Bach Dinh Le, Ming Li, David Lo

Research Collection School Of Computing and Information Systems

Tools that alert developers about library vulnerabilities depend on accurate, up-to-date vulnerability databases which are maintained by security researchers. These databases record the libraries related to each vulnerability. However, the vulnerability reports may not explicitly list every library and human analysis is required to determine all the relevant libraries. Human analysis may be slow and expensive, which motivates the need for automated approaches. Researchers and practitioners have proposed to automatically identify libraries from vulnerability reports using extreme multi-label learning (XML). While state-of-the-art XML techniques showed promising performance, their experimental settings do not practically fit what happens in reality. Previous studies …