Computer Sciences Commons^™

Efficient Anomaly Detection Driven By Different Machine Learning Architectures And Models, William Marfo

Open Access Theses & Dissertations

The rapid growth and ubiquitous adoption of the internet and cyber-physical systems (CPS) have fundamentally transformed modern communication, work, and human-system interactions. While networks now form the backbone of critical digital ecosystems, enabling seamless data transmission across diverse, interconnected systems, this increased connectivity also expands the attack surface, making real-time detection of network intrusions and anomalies a pressing challenge. Detecting unusual activities within network infrastructure requires advanced data traffic analysis to differentiate between legitimate and malicious interactions. Traditional approaches to network anomaly detectionâ??such as rule-based and signature-based systemsâ??often depend on predefined patterns to identify known anomalies, limiting their effectiveness against …

Lova3 : Learning To Visual Question Answering, Asking And Assessment, Henry Hengyuan Zhao, Pan Zhou, Difei Gao, Bai Shou, Mike Zheng Shou

Research Collection School Of Computing and Information Systems

Question answering, asking, and assessment are three innate human traits crucial for understanding the world and acquiring knowledge. By enhancing these capabilities, humans can more effectively utilize data, leading to better comprehension and learning outcomes. Current Multimodal Large Language Models (MLLMs) primarily focus on question answering, often neglecting the full potential of questioning and assessment skills. Inspired by the human learning mechanism, we introduce LOVA3 , an innovative framework named “Learning tO Visual question Answering, Asking and Assessment,” designed to equip MLLMs with these additional capabilities. Our approach involves the creation of two supplementary training tasks GenQA and EvalQA, aiming …

Sampdetox : Black-Box Backdoor Defense Via Perturbation-Based Sample Detoxification, Yanxin Yang, Chentao Jia, Dengke Yan, Ming Hu, Tianlin Li, Xiaofei Xie, Xian Wei, Mingsong Chen

Research Collection School Of Computing and Information Systems

The advancement of Machine Learning has enabled the widespread deployment of Machine Learning as a Service (MLaaS) applications. However, the untrustworthy nature of third-party ML services poses backdoor threats. Existing defenses in MLaaS are limited by their reliance on training samples or white-box model analysis, highlighting the need for a black-box backdoor purification method. In our paper, we attempt to use diffusion models for purification by introducing noise in a forward diffusion process to destroy backdoors and recover clean samples through a reverse generative process. However, since a higher noise also destroys the semantics of the original samples, it still …

Exploring Post-Covid-19 Health Effects And Features With Advanced Machine Learning Techniques, Muhammad N. Islam, Md S. Islam, Nahid H. Shourav, Iftiaqur Rahman, Faiz A. Faisal, Md M. Islam, Iqbal H. Sarker

Research outputs 2022 to 2026

COVID-19 is an infectious respiratory disease that has had a significant impact, resulting in a range of outcomes including recovery, continued health issues, and the loss of life. Among those who have recovered, many experience negative health effects, particularly influenced by demographic factors such as gender and age, as well as physiological and neurological factors like sleep patterns, emotional states, anxiety, and memory. This research aims to explore various health factors affecting different demographic profiles and establish significant correlations among physiological and neurological factors in the post-COVID-19 state. To achieve these objectives, we have identified the post-COVID-19 health factors and …

Sustainable Energysense: A Predictive Machine Learning Framework For Optimizing Residential Electricity Consumption, Murad Al-Rajab, Samia Loucif

All Works

In a world where electricity is often taken for granted, the surge in consumption poses significant challenges, including elevated CO2 emissions and rising prices. These issues not only impact consumers but also have broader implications for the global environment. This paper endeavors to propose a smart application dedicated to optimizing the electricity consumption of household appliances. It employs Augmented Reality (AR) technology along with YOLO to detect electrical appliances and provide detailed electricity consumption insights, such as displaying the appliance consumption rate and computing the total electricity consumption based on the number of hours the appliance was used. The application …

Artificial Intelligence And Machine Learning In Cancer Pain: A Systematic Review, Vivian Salama, Brandon Godinich, Yimin Geng, Laia Humbert-Vidan, Laura Maule, Kareem A Wahid, Mohamed A Naser, Renjie He, Abdallah S R Mohamed, Clifton D Fuller, Amy C Moreno

Faculty, Staff and Student Publications

Background/objectives: Pain is a challenging multifaceted symptom reported by most cancer patients. This systematic review aims to explore applications of artificial intelligence/machine learning (AI/ML) in predicting pain-related outcomes and pain management in cancer.

Methods: A comprehensive search of Ovid MEDLINE, EMBASE and Web of Science databases was conducted using terms: "Cancer," "Pain," "Pain Management," "Analgesics," "Artificial Intelligence," "Machine Learning," and "Neural Networks" published up to September 7, 2023. AI/ML models, their validation and performance were summarized. Quality assessment was conducted using PROBAST risk-of-bias andadherence to TRIPOD guidelines.

Results: Forty four studies from 2006 to 2023 were included. Nineteen studies used …

Safety Through Feedback In Constrained Rl, Shashank Reddy Chirra, Pradeep Varakantham, Praveen Paruchuri

Research Collection School Of Computing and Information Systems

In safety-critical RL settings, the inclusion of an additional cost function is often favoured over the arduous task of modifying the reward function to ensure the agent's safe behaviour. However, designing or evaluating such a cost function can be prohibitively expensive. For instance, in the domain of self-driving, designing a cost function that encompasses all unsafe behaviours (e.g., aggressive lane changes, risky overtakes) is inherently complex, it must also consider all the actors present in the scene making it expensive to evaluate. In such scenarios, the cost function can be learned from feedback collected offline in between training rounds. This …

Decoding Neural Networks: An Information-Theoretic Guide To Interpretability, Error Analysis And Efficiency, Mackenzie J. Meni

Theses and Dissertations

This dissertation addresses critical challenges in neural network design by leveraging entropy-based techniques to improve model efficiency, interpretability, and bias reduction. Focusing on the unique demands of computer vision applications, particularly object detection and classification for real-time systems, this work introduces a series of innovative methods centered on information theory. At the core of these methods is the Probabilistic Explanations of Entropic Knowledge (PEEK) framework, a tool developed to analyze and visualize entropy distributions across feature maps. PEEK offers insights into information flow within neural networks, making it possible to pinpoint layers that contribute meaningfully to decision-making or identify those …

Unified Generative And Discriminative Training For Multi-Modal Large Language Models, Wei Chow, Juncheng Li, Kaihang Pan, Qifan Yu, Hao Fei, Zhiqi Ge, Shuai Yang, Siliang Teng, Hanwang Zhang, Qianru Sun

Research Collection School Of Computing and Information Systems

In recent times, Vision-Language Models (VLMs) have been trained under two predominant paradigms. Generative training has enabled Multimodal Large Language Models (MLLMs) to tackle various complex tasks, yet issues such as hallucinations and weak object discrimination persist. Discriminative training, exemplified by models like CLIP, excels in zero-shot image-text classification and retrieval, yet struggles with complex scenarios requiring fine-grained semantic differentiation. This paper addresses these challenges by proposing a unified approach that integrates the strengths of both paradigms. Considering interleaved image-text sequences as the general format of input samples, we introduce a structure-induced training strategy that imposes semantic relationships between input …

Designing Customized Loss Functions For Training Deep Neural Networks, Ali Pourramezan Fard

Electronic Theses and Dissertations

This dissertation explores the critical role of loss functions in enhancing the predictive performance of deep machine learning models. Loss functions are an integral element of all the ongoing advances we witness daily in this domain. I design custom loss functions and their impacts on various machine learning tasks, particularly in computer vision.

In the first stage of my research, I aim to improve the prediction performance of deep learning models by providing them with more precise feedback associated with task requirements. This led me to create the concept of assistive loss functions. My first proposed loss function, inspired by …

Applying Machine‐Learning Methods To Laser Acceleration Of Protons: Lessons Learned From Synthetic Data, Ronak Desai, Thomas Zhang, J. J. Felice, Ricky Oropeza, Joseph R. Smith, Alona Kryshchenko, Chris Orban, Michael L. Dexter, Anil K. Patnaik

Faculty Publications

In this study, we consider three different machine-learning methods—a three-hidden-layer neural network, support vector regression, and Gaussian process regression—and compare how well they can learn from a synthetic data set for proton acceleration in the Target Normal Sheath Acceleration regime. The synthetic data set was generated from a previously published theoretical model by Fuchs et al. 2005 that we modified. Once trained, these machine-learning methods can assist with efforts to maximize the peak proton energy, or with the more general problem of configuring the laser system to produce a proton energy spectrum with desired characteristics. In our study, we focus …

Machine Learning-Driven Process Analysis And Optimization In Solid-State Welding And Fusion-Based Additive Manufacturing, Radif Uddin Ahmed

Master's Theses

In the modern era of advanced manufacturing, optimizing process parameters is pivotal in ensuring the quality and reliability of sophisticated component fabrication. This study presents a novel, data-driven approach to parameter optimization in two cutting-edge manufacturing techniques: Friction Stir Welding (FSW) and Laser Powder Bed Fusion (LPBF). By leveraging machine learning methodologies, this research addresses the critical challenge of efficiently determining optimal process parameters, a task traditionally relying on time-consuming and resource-intensive trial-and-error methods. This study will lead to a robust data-driven framework for process analysis of more advanced manufacturing techniques like the Additive Friction Stir Deposition (AFSD) process. Friction …

Predicting And Monitoring Immune Checkpoint Inhibitor Therapy Using Artificial Intelligence In Pancreatic Cancer, Guangbo Yu, Zigeng Zhang, Aydin Eresen, Qiaoming Hou, Farideh Amirrad, Sha Webster, Surya M. Nauli, Vahid Yaghmai, Zhuoli Zhang

Pharmacy Faculty Articles and Research

Pancreatic cancer remains one of the most lethal cancers, primarily due to its late diagnosis and limited treatment options. This review examines the challenges and potential of using immunotherapy to treat pancreatic cancer, highlighting the role of artificial intelligence (AI) as a promising tool to enhance early detection and monitor the effectiveness of these therapies. By synthesizing recent advancements and identifying gaps in the current research, this review aims to provide a comprehensive overview of how AI and immunotherapy can be integrated to develop more personalized and effective treatment strategies. The insights from this review may guide future research efforts …

Enhancedbert: A Python Software Tailored For Arabic Word Sense Disambiguation, Sanaa Kaddoura, Reem Nassar

All Works

EnhancedBERT is a software framework designed to disambiguate Arabic polysemous terms using advanced natural language processing techniques. It integrates transformer architectures with ensemble methods to achieve high performance in understanding and processing Arabic text. The framework provides a flexible pipeline that can be directly utilized or fine-tuned according to specific needs. EnhancedBERT stands out for its ease of use, leveraging transformer-based models combined with ensemble strategies to provide superior contextual understanding. This contextual awareness makes it an invaluable tool for researchers and practitioners tackling complexities in Arabic language processing.

A Large Scale Multi Institutional Study For Radiomics Driven Machine Learning For Meningioma Grading, Mert Karabacak, Shiv Patil, Rui Feng, Raj K. Shrivastava, Konstantinos Margetis

Department of Medicine Faculty Papers

This study aims to develop and evaluate radiomics-based machine learning (ML) models for predicting meningioma grades using multiparametric magnetic resonance imaging (MRI). The study utilized the BraTS-MEN dataset's training split, including 698 patients (524 with grade 1 and 174 with grade 2-3 meningiomas). We extracted 4872 radiomic features from T1, T1 with contrast, T2, and FLAIR MRI sequences using PyRadiomics. LASSO regression reduced features to 176. The data was split into training (60%), validation (20%), and test (20%) sets. Five ML algorithms (TabPFN, XGBoost, LightGBM, CatBoost, and Random Forest) were employed to build models differentiating low-grade (grade 1) from high-grade …

Radiomics-Based Machine Learning With Natural Gradient Boosting For Continuous Survival Prediction In Glioblastoma, Mert Karabacak, Shiv Patil, Zachary C. Gersey, Ricardo J. Komotar, Konstantinos Margetis

SKMC Student Presentations and Publications

(1) Background: Glioblastoma (GBM) is the most common primary malignant brain tumor in adults, with an aggressive disease course that requires accurate prognosis for individualized treatment planning. This study aims to develop and evaluate a radiomics-based machine learning (ML) model to estimate overall survival (OS) for patients with GBM using pre-treatment multi-parametric magnetic resonance imaging (MRI). (2) Methods: The MRI data of 865 patients with GBM were assessed, comprising 499 patients from the UPENN-GBM dataset and 366 patients from the UCSF-PDGM dataset. A total of 14,598 radiomic features were extracted from T1, T1 with contrast, T2, and FLAIR MRI sequences …

Artificial Intelligence And Machine Learning In Ocular Oncology, Retinoblastoma (Armor): Experience With A Multiracial Cohort, Vijitha S. Vempuluru, Rajiv Viriyala, Virinchi Ayyagari, Komal Bakal, Patanjali Bhamidipati, Krishna Krishore Dhara, Sandor R. Ferenczy, Carol L. Shields, Swathi Kaliki

Wills Eye Hospital Papers

Background: The color variation in fundus images from differences in melanin concentrations across races can affect the accuracy of artificial intelligence and machine learning (AI/ML) models. Hence, we studied the performance of our AI model (with proven efficacy in an Asian-Indian cohort) in a multiracial cohort for detecting and classifying intraocular RB (iRB). Methods: Retrospective observational study. Results: Of 210 eyes, 153 (73%) belonged to White, 37 (18%) to African American, 9 (4%) to Asian, 6 (3%) to Hispanic races, based on the U.S. Office of Management and Budget's Statistical Policy Directive No.15 and 5 (2%) had no reported race. …

Predictive Residual Neural Networks For Optical Trapping Of Small Particles, Nasim Mohammadi Estrakhri, Ponthea Zahraii, Saman Kashanchi, Nooshin M. Estakhri

Engineering Faculty Articles and Research

Optical tweezers provide a non-contact method to trap, move, and manipulate micro- and nano-sized objects. Using properly designed dielectric and plasmonic nanostructure configurations, optical tweezers have been tailored to create stable and precise trapping for nanoscale objects. Recent advances in numerical optimization techniques allow further enhancement in nanoscale optical traps through inverse optimization of such configurations. One of the main challenges in such optimization approaches is the time-consuming nature of full-wave simulation of nanostructures and postprocessing steps to extract optical forces. To address this challenge, we introduce a surrogate solver based on residual neural networks that can accurately predict the …

Data Provenance Via Differential Auditing, Xin Mu, Ming Pang, Feida Zhu

Research Collection School Of Computing and Information Systems

With the rising awareness of data assets, data governance, which is to understand where data comes from, how it is collected, and how it is used, has been assuming evergrowing importance. One critical component of data governance gaining increasing attention is auditing machine learning models to determine if specific data has been used for training. Existing auditing techniques, like shadow auditing methods, have shown feasibility under specific conditions such as having access to label information and knowledge of training protocols. However, these conditions are often not met in most real-world applications. In this paper, we introduce a practical framework for …

Optimizing Transport Predictive Modeling With Simulation-Based Statistical Inference, Quyen Tran '27, Mamunur Rashid

Annual Student Research Poster Session

Simulation-based statistical inference (SBI) leverages computer simulations to help scientists understand and analyze complex data. This project explores how SBI techniques can be used to analyze transportation data. We use modern computational methods, including machine learning models, to improve the accuracy of predictions and decision-making in transportation planning. Our study focuses on applying two SBI methods, Approximate Bayesian Computation - Markov Chain Monte Carlo and Synthetic Likelihood, to create synthetic data for training machine learning models. These models show the potential of SBI to handle uncertain data. It also highlights the practical benefits of SBI in making predictions and decisions …

Improving Out-Of-Distribution Detection With Disentangled Foreground And Background Features, Choubo Ding, Guansong Pang

Research Collection School Of Computing and Information Systems

Detecting out-of-distribution (OOD) inputs is a principal task for ensuring the safety of deploying deep-neural-network classifiers in open-set scenarios. OOD samples can be drawn from arbitrary distributions and exhibit deviations from in-distribution (ID) data in various dimensions, such as foreground features (e.g., objects in CIFAR100 images vs. those in CIFAR10 images) and background features (e.g., textural images vs. objects in CIFAR10). Existing methods can confound foreground and background features in training, failing to utilize the background features for OOD detection. This paper considers the importance of feature disentanglement in out-of-distribution detection and proposes the simultaneous exploitation of both foreground and …

Promise And Peril Of Collaborative Code Generation Models : Balancing Effectiveness And Memorization, Zhi Chen, Lingxiao Jiang

Research Collection School Of Computing and Information Systems

In the rapidly evolving field of machine learning, training models with datasets from various locations and organizations presents significant challenges due to privacy and legal concerns. The exploration of effective collaborative training settings, which are capable of leveraging valuable knowledge from distributed and isolated datasets, is increasingly crucial. This study investigates key factors that impact the effectiveness of collaborative training methods in code next-token prediction, as well as the correctness and utility of the generated code, showing the promise of such methods. Additionally, we evaluate the memorization of different participant training data across various collaborative training settings, including centralized, federated, …

Can Federated Learning Solve Ai’S Data Privacy Problem?: A Legal Analysis, Warren B. Chik, Florian Gamper

Research Collection Yong Pung How School Of Law

Federated learning (FL) is a method of training AI systems on different datasets without sharing data. The promise of FL is to enable AI systems to be trained on data, including personal data, while preserving data privacy and confidentiality, and thus, inter alia, facilitate compliance with data protection legislation. FL has generated a considerable interest amongst the computer science community, yet there is a dearth of legal analysis of FL. This is a problem because the question of whether FL facilitates compliance with data protection legislation is a legal question. This article will fill this lacuna by providing a comprehensive …

Etdsuite: A Toolkit To Mine Electronic Theses And Dissertations To Enrich Scholarly Big Data Using Natural Language Processing And Computer Vision, Muntabir Hasan Choudhury

Computer Science Theses & Dissertations

In the past decades, there has been a growing interest in mining scientific documents to obtain domain knowledge automatically. One of the understudied types of scientific documents is Electronic Theses and Dissertations (ETDs), as ETDs have distinct features compared with conference proceedings and journal articles. ETDs usually serve as partial requirements of academic degrees for students pursuing higher education. They are book-length documents (i.e., 100 – 400 pages long), and the topics may shift across chapters, exhibit the significant contribution of a student’s research over the entire degree pursuing period, and have unique metadata schema and page layouts. However, the …

Using Machine Learning And Deep Learning Algorithms To Improve Low Birthweight Prediction, Yang Ren

Theses and Dissertations

Low birthweight (LBW) is a major public health issue resulting in increased neonatal mortality and long-term health complications. Traditional LBW analysis methods, focusing on incidence rates and risk factors through statistical models, often struggle with complex unseen data, and thus, their effectiveness is limited in early prevention of LBW, requiring more advanced LBW prediction models. Therefore, this dissertation delves into this important research area by proposing and examining novel machine learning (ML) and deep learning (DL) algorithms, aiming to predict LBW more accurately during the early stage of pregnancy. This dissertation consists of three studies, strategically designed to build upon …

A Guideline For Open-Source Tools To Make Medical Imaging Data Ready For Artificial Intelligence Applications: A Society Of Imaging Informatics In Medicine (Siim) Survey, Sanaz Vahdati, Bardia Khosravi, Elham Mahmoudi, Kuan Zhang, Pouria Rouzrokh, Shahriar Faghani, Mana Moassefi, Aylin Tahmasebi, Katherine Andriole, Peter Chang, Keyvan Farahani, Mona Flores, Les Folio, Sina Houshmand, Maryellen Giger, Judy Gichoya, Bradley Erickson

Department of Radiology Faculty Papers

In recent years, the role of Artificial Intelligence (AI) in medical imaging has become increasingly prominent, with the majority of AI applications approved by the FDA being in imaging and radiology in 2023. The surge in AI model development to tackle clinical challenges underscores the necessity for preparing high-quality medical imaging data. Proper data preparation is crucial as it fosters the creation of standardized and reproducible AI models while minimizing biases. Data curation transforms raw data into a valuable, organized, and dependable resource and is a fundamental process to the success of machine learning and analytical projects. Considering the plethora …

A Generalized Machine Learning Model For Long-Term Coral Reef Monitoring In The Red Sea, Justin J. Gapper, Surendra Maharjan, Wenzhao Li, Erik Linstead, Surya Prakash Tiwari, Mohamed A. Qurban, Hesham El-Askary

Mathematics, Physics, and Computer Science Faculty Articles and Research

Coral reefs, despite covering less than 0.2 % of the ocean floor, harbor approximately 35 % of all known marine species, making their conservation critical. However, coral bleaching, exacerbated by climate change and phenomena such as El Niño, poses a significant threat to these ecosystems. This study focuses on the Red Sea, proposing a generalized machine learning approach to detect and monitor changes in coral reef cover over an 18-year period (2000–2018). Using Landsat 7 and 8 data, a Support Vector Machine (SVM) classifier was trained on depth-invariant indices (DII) derived from the Gulf of Aqaba and validated against ground …

Attention-Based Load Forecasting With Bidirectional Finetuning, Firuz Kamalov, Inga Zicmane, Murodbek Safaraliev, Linda Smail, Mihail Senyuk, Pavel Matrenin

All Works

Accurate load forecasting is essential for the efficient and reliable operation of power systems. Traditional models primarily utilize unidirectional data reading, capturing dependencies from past to future. This paper proposes a novel approach that enhances load forecasting accuracy by fine tuning an attention-based model with a bidirectional reading of time-series data. By incorporating both forward and backward temporal dependencies, the model gains a more comprehensive understanding of consumption patterns, leading to improved performance. We present a mathematical framework supporting this approach, demonstrating its potential to reduce forecasting errors and improve robustness. Experimental results on real-world load datasets indicate that our …

The Evaluation Of Machine Learning Techniques For Isotope Identification Contextualized By Training And Testing Spectral Similarity, Aaron P. Fjelsted, Tyler J. Morrow, Clayton D. Scott, Yilun Zhu, Darren E. Holland, Azaree T. Lintereur, Douglas E. Wolfe

Faculty Publications

Precise gamma-ray spectral analysis is crucial in high-stakes applications, such as nuclear security. Research efforts toward implementing machine learning (ML) approaches for accurate analysis are limited by the resemblance of the training data to the testing scenarios. The underlying spectral shape of synthetic data may not perfectly reflect measured configurations, and measurement campaigns may be limited by resource constraints. Consequently, ML algorithms for isotope identification must maintain accurate classification performance under domain shifts between the training and testing data. To this end, four different classifiers (Ridge, Random Forest, Extreme Gradient Boosting, and Multilayer Perceptron) were trained on the same dataset …

Regulating Algorithmic Harms, Sylvia Lu

Law & Economics Working Papers

In recent years, the rapid expansion of artificial intelligence (AI) innovations has led to a rise in algorithmic harms—harms emerging from AI operations that pose significant threats to civil rights and democratic values in today’s technological landscape. A facial recognition system for improving criminal detection wrongly collected sensitive personal data and flagged racial minorities as shoplifters. A risk-prediction algorithm adopted to identify patients denied medical treatment to Black individuals with poor health conditions. A social media algorithm intended to boost social engagement exacerbated addictive behavior and mental illness in teenagers. These harms are becoming increasingly ubiquitous yet often manifest in …