Digital Commons Network^™

Reliable-Data-Split (Rds): Maximizing Model Potential With Reinforced Selection Strategy, Hoang D. Nguyen, Xuan-Son Vu, Quoc Tuan Truong, Duc-Trong Le

Research Collection School Of Computing and Information Systems

The nexus between data characteristics and parametric models is fundamental for developing effective and reliable artificial intelligence (AI) systems. Mismatches in data properties for model development may lead to deleterious effects on AI model performance in machine learning practice. This paper proposes a Reliable Data Split (RDS) procedure to learn how to select data points that will generalise the target domain adequately by employing prior knowledge of the data generative process. We introduce a reinforced selection strategy using deep reinforcement learning with diverse black box predictors in maximising ensemble rewards as the proxy of model performance potential while maintaining an …

A Learning‑Augmented Dynamic Programming Approach For Orienteering Problem With Time Windows, Guansheng Peng, Lining Xing, Fuyan Song Ma, Aldy Gunawan, Aldy Gunawan

Research Collection School Of Computing and Information Systems

Recent years have witnessed a surge of interest in solving combinatorial optimization problems (COPs) using machine learning techniques. Motivated by this trend, we propose a learning-augmented exact approach for tackling an NP-hard COP, the Orienteering Problem with Time Windows, which aims to maximize the total score collected by visiting a subset of vertices in a graph within their time windows. Traditional exact algorithms rely heavily on domain expertise and meticulous design, making it hard to achieve further improvements. By leveraging deep learning models to learn effective relaxations of problem restrictions from data, our approach enables significant performance gains in an …

Semi‑Supervised Graph Anomaly Detection Via Robust Homophily Learning, Guoguo Ai, Hezhe Qiao, Hui Yan, Guansong Pang

Research Collection School Of Computing and Information Systems

Current semi-supervised graph anomaly detection (GAD) methods utilizes a small set of labeled normal nodes to identify abnormal nodes from a large set of unlabeled nodes in a graph. These methods posit that 1) normal nodes share a similar level of homophily and 2) the labeled normal nodes can well represent the homophily patterns in the entire normal class. However, this assumption often does not hold well since normal nodes in a graph can exhibit diverse homophily in real-world GAD datasets. In this paper, we propose RHO, namely Robust Homophily Learning, to adaptively learn such homophily patterns. RHO consists of …

Marsanywhere: Dataset And Cross-View Diffusion Model For Satellite-To-Ground View Synthesis With Mars Data, Benjamin T. Hinchliff

Master's Theses

Satellite-to-ground view synthesis aims to create a realistic ground view image from a corresponding satellite view image. This is a well-studied problem for street level imagery, with good results being achieved by using modern image synthesis techniques such as diffusion models. However, despite the public availability of satellite and ground level imagery on Mars, these techniques have yet to be applied to the domain due to difficulties in collating and processing the data into a usable form. We address this deficiency by creating a dataset consisting of ground view panorama imagery from the Perseverance rover, along with associated satellite view …

Curvilinear Image Segmentation Using Multiscale Variational U-Net, Rebekah Fortes

LSU New Orleans Theses and Dissertations

Segmentation of curvilinear structures such as water contours, cracks in cement, and vascular networks in biomedical imaging, poses unique challenges due to extreme class imbalance, irregular morphology, low contrast against complex backgrounds, and the need to preserve global connectivity while detecting fine-scale details. We propose a Multiscale Variational U-Net (MSVU-Net) architecture designed specifically to address these challenges. The model integrates multiscale convolutional filters to capture both global context and local detail, while embedding a variational model in the bottleneck layer to enhance structural representation. To mitigate class imbalance and improve fidelity, the network optimizes a hybrid loss function that combines …

Topic Modeling And Culturomic Analysis Of 30,000 Books Over 100 Years Using Gensim, Michael A. Freeman

Electronic Theses and Dissertations

This thesis explores the cultural influence of historical events on English-language fiction published between 1820 and 1929. Using a corpus of 30,256 digitized books from Project Gutenberg, Latent Dirichlet Allocation (LDA) topic modeling was applied to identify recurring themes across eleven decades. The study sought to determine whether historically significant events could be detected within fictional narratives. One clear instance emerged: Napoleon Bonaparte and the Napoleonic Wars appeared explicitly in the 1820s corpus. Beyond this, several thematic patterns were observed—such as maritime language in the 1840s, national identity in the 1880s, and youth-oriented dialogue in the early 20th century—that plausibly …

Understanding Bias And Fairness In Large Language Models: An Empirical Study, Joshua Johnson

Electrical Engineering and Computer Science Undergraduate Honors Theses

This thesis investigates demographic bias in large language models (LLMs) through the use of evaluating outcome disparities when utilized in decision making tasks as well as underlying associations that could contribute to furthering these disparities. Using profiles from the Adult dataset, we analyze how Gemini 2.0 Flash performs in an income prediction task using zero-shot and few-shot prompting methods. Our findings show that models exhibit measurable differences in demographic parity and false positive rates, with the use of few-shot prompting reducing these disparities. Alongside this line of testing, we tested associational bias in Qwen 2.5 using probability based association tests …

Definition Of The 3d Position And Motion Status Of The Moving Heart Based On 2d Projections, Lawrence D. Orijuela

Electronic Theses, Projects, and Dissertations

This thesis presents a novel application of deep learning to the estimation of pulmonary vein coordinates using X-ray image pairs from a FORBILD Thorax phantom derived motion dataset. A Siamese neural network was developed to predict the 3D coordinates of one pulmonary vein at a time, specifically the Right Superior Pulmonary Vein (RSPV), Left Superior Pulmonary Vein (LSPV), Left Inferior Pulmonary Vein (LIPV), or Right Inferior Pulmonary Vein (RIPV), based on two-dimensional projection images.

The input data consisted of over 1.6 million grayscale X-ray image pairs across 1331 virtual patients, each annotated with ground truth 3D coordinates. To manage memory …

The Future Is Now: Empowering Society Through Ai Literacy, Jason S. Wrench, Sanae Elmoudden

Milne Open Textbooks

Artificial Intelligence (AI) is no longer a futuristic concept—it is the reality of the present. From the algorithms shaping our social media feeds to the generative tools transforming our workplaces, AI has permeated every aspect of modern life. The Future is Now moves beyond the hype to provide a comprehensive roadmap for understanding, navigating, and shaping this technological revolution.

Demystifying the Machine

This textbook serves as a user-friendly guide to the “black box” of AI. It breaks down complex technical concepts—from machine learning and neural networks to large language models—making them accessible to students across all disciplines. By establishing a …

Sopo: Text-To-Motion Generation Using Semi-Online Preference Optimization, Xiaofeng Tan, Hongsong Wang, Xin Geng, Pan Zhou

Research Collection School Of Computing and Information Systems

Text-to-motion generation is essential for advancing the creative industry but often presents challenges in producing consistent, realistic motions. To address this, we focus on fine-tuning text-to-motion models to consistently favor highquality, human-preferred motions—a critical yet largely unexplored problem. In this work, we theoretically investigate the DPO under both online and offline settings, and reveal their respective limitation: overfitting in offline DPO, and biased sampling in online DPO. Building on our theoretical insights, we introduce Semi-online Preference Optimization (SoPo), a DPO-based method for training text-to-motion models using “semi-online” data pair, consisting of unpreferred motion from online distribution and preferred motion in …

Uniteformer: Unifying Node And Edge Modalities In Transformers For Vehicle Routing Problem, Dian Meng, Zhiguang Cao, Jie Gao, Yaoxin Wu, Yaqing Hou

Research Collection School Of Computing and Information Systems

Neural solvers for the Vehicle Routing Problem (VRP) have typically relied on either node or edge inputs, limiting their flexibility and generalization in real-world scenarios. We propose UniteFormer, a unified neural solver that supports node-only, edge-only, and hybrid input types through a single model trained via joint edge-node modalities. UniteFormer introduces: (1) a mixed encoder that integrates graph convolutional networks and attention mechanisms to collaboratively process node and edge features, capturing cross-modal interactions between them; and (2) a parallel decoder enhanced with query mapping and a feed-forward layer for improved representation. The model is trained with REINFORCE by randomly sampling …

Learning Memory-Enhanced Improvement Heuristics For Flexible Job Shop Scheduling, Jiaqi Wang, Zhiguang Cao, Peng Zhao, Rui Cao, Yubin Xiao, Yuan Jiang, You Zhou

Research Collection School Of Computing and Information Systems

The rise of smart manufacturing under Industry 4.0 introduces mass customization and dynamic production, demanding more advanced and flexible scheduling techniques. The flexible job-shop scheduling problem (FJSP) has attracted significant attention due to its complex constraints and strong alignment with real-world production scenarios. Current deep reinforcement learning (DRL)-based approaches to FJSP predominantly employ constructive methods. While effective, they often fall short of reaching (near-)optimal solutions. In contrast, improvement-based methods iteratively explore the neighborhood of initial solutions and are more effective in approaching optimality. However, the flexible machine allocation in FJSP poses significant challenges to the application of this framework, including …

Instance-Level Video Depth In Groups Beyond Occlusions, Yuan Liang, Yang Zhou, Ziming Sun, Tianyi Xiang, Guiqing Li, Shengfeng He

Research Collection School Of Computing and Information Systems

Depth estimation in dynamic, multi-object scenes remains a major challenge, especially under severe occlusions. Existing monocular models, including foundation models, struggle with instance-wise depth consistency due to their reliance on global regression. We tackle this problem from two key aspects: data and methodology. First, we introduce the Group Instance Depth (GID) dataset, the first large-scale video depth dataset with instance-level annotations, featuring 101,500 frames from real-world activity scenes. GID bridges the gap between synthetic and real-world depth data by providing high-fidelity depth supervision for multi-object interactions. Second, we propose InstanceDepth, the first occlusion-aware depth estimation framework for multi-object environments. Our …

Robust Hallucination Detection In Llms Via Adaptive Token Selection, Mengjia Niu, Hamed Haddadi, Guansong Pang

Research Collection School Of Computing and Information Systems

Hallucinations in large language models (LLMs) pose significant safety concerns that impede their broader deployment. Recent research in hallucination detection has demonstrated that LLMs’ internal representations contain truthfulness hints, which can be harnessed for detector training. However, the performance of these detectors is heavily dependent on the internal representations of predetermined tokens, fluctuating considerably when working on free-form generations with varying lengths and sparse distributions of hallucinated entities. To address this, we propose HaMI, a novel approach that enables robust detection of hallucinations through adaptive selection and learning of critical tokens that are most indicative of hallucinations. We achieve this …

Sempo: Lightweight Foundation Models For Time Series Forecasting, Hui He, Kun Yi, Yuanchi Ma, Qi Zhang, Zhengdong Niu, Guansong Pang

Research Collection School Of Computing and Information Systems

The recent boom of large pre-trained models witnesses remarkable success in developing foundation models (FMs) for time series forecasting. Despite impressive performance across diverse downstream forecasting tasks, existing time series FMs possess massive network architectures and require substantial pre-training on large-scale datasets, which significantly hinders their deployment in resource-constrained environments. In response to this growing tension between versatility and affordability, we propose SEMPO, a novel lightweight foundation model that requires pretraining on relatively small-scale data, yet exhibits strong general time series forecasting. Concretely, SEMPO comprises two key modules: 1) energy-aware SpEctral decomposition module, that substantially improves the utilization of pre-training …

Examining The Roles Of Embodiment And Theory Of Mind In Shaping User Perceptions Of Llm-Driven Conversational Agents, Elizabeth A. Schlesener

All Dissertations

Large Language Models (LLMs) have advanced conversational agents, enabling natural, human-like interactions in domains such as education, programming, and workplace collaboration. Yet, user distrust persists over privacy, accuracy, and bias. As developers work to mitigate these issues and human-AI collaboration expands, reinforcing trust in LLM-driven systems is essential. To address this problem, this dissertation explores the role of anthropomorphic form in LLM-driven conversational agents and its impact on user perception.

According to the familiarity thesis, humans attribute human-like characteristics to nonhuman entities — a process known as anthropomorphism — to better comprehend unfamiliar phenomena, based on the assumption that they …

Ai In Consideration Of Her: Accounting For Gendered Workplace Dynamics In The Design And Evaluation Of Human-Centered Ai Integration In Everyday Workplaces, Kelsea S. Schulenberg

All Dissertations

Rapid advancements in the technical capabilities and availability of generative Artificial Intelligence (AI) systems, such as OpenAI's ChatGPT, have drawn widespread attention to the opportunities and challenges associated with AI integration into everyday workplaces (i.e., office-type work). Following calls for organizations to consider the ethical and workplace-specific impacts of generative AI's use before integrating it into the workplace, this dissertation addresses three critical gaps in Human-Centered Computing (HCC) and AI workplace integration research. First, this dissertation unpacks the underdeveloped links between women's representation - or lack thereof - in AI-related fields and how their experiences with gendered workplace dynamics in …

From Morphology To Machine Learning And Genomics: Understanding Phenotypic Variation In Wild Ducks, Sara Gonzalez

Open Access Theses & Dissertations

Understanding the genetic underpinning and distribution of phenotypic variation within and between divergent groups is core towards shedding light into how populations diverge and adapt, as well as how hybridization breaks or builds on these scenarios; and thus, central to evolutionary biology. In wild organisms, however, quantifying and linking phenotypic traits to underlying genetic processes, like mutation, gene expression, epigenetics and allele interactions, remains challenging. This difficulty arises from the complex interplay between morphology, environment, and gene regulation, as well as the logistical barriers of collecting and standardizing large-scale data across individuals and populations. As a result, researchers are increasingly …

Physics-Guided Strategies For Enhancing Neural Networks Trained With Limited Data, Jose Guadalupe Perez Zamora

Open Access Theses & Dissertations

Deep neural networks excel at a wide range of processing tasks across various disciplines. However, the quantity and quality of data significantly impact network performance. In specialized domains, high-quality datasets are often difficult to gather, interpret, and curate for effective learning. Few-shot learning techniques, including transfer learning, data augmentation, and meta-learning, have emerged to address these constraints.

We propose three physics-guided strategies for enhancing neural networks trained with limited data: (1) combining existing models like LSTMs with Physics-Informed Neural Networks through two-branch architectures that merge their outputs, (2) deriving custom physics-informed data augmentation algorithms to expand limited datasets, and (3) …

Exploiting The In-Distribution Embedding Space With Deep Learning And Gaussian Discriminant Analysis For An Out-Of-Distribution Malware Attach Detection, Tosin Olusola Ige

Open Access Theses & Dissertations

State-of-the-art machine and deep learning models generally perform well on previously seen data, albeit with wrong close world assumption that all real-world data are from previously seen train and validation samples, hence there poor performance when exposed to data which deviates from previously seen training and validation set. This is clearly evident in the domain of cybersecurity where the world continues to experience several high profile malware attacks despite advancement in state-of-the-art research. The reason being that the constant evolvement of innovation in the development of tools and method deployed to carry out various attacks had given hackers and other …

Hubert-Based Models And Evaluation Strategies For Pragmatically-Faithful Speech To Speech Translation, Javier Vazquez

Open Access Theses & Dissertations

Pragmatic fidelity in speech-to-speech translation (S2ST) has largely been understudied, leading to communication tools inadequate to support non-superficial dialog. We aim to improve pragmatic faithfulness in English-Spanish translation through the development of machine learning models that are able to predict a corresponding pragmatic representation in the other language. To evaluate performance, we developed a pipeline that utilizes a recently-developed pragmatic similarity evaluation metric to compare models. Further, we developed models that exploit HuBERT features as these have been found suitable for various prosody and pragmatics related tasks. Our models outperformed human and state-of-the-art predictions, albeit the methodology being limited to …

Synthetic Dataset For Understanding Negation In Text-Guided Image Editing, Nhat-Tan Bui

Graduate Theses and Dissertations

Negation is a fundamental linguistic concept used by humans to convey information that they do not desire. Despite this, minimal research has focused on negation within text-guided image editing. This lack of research means that vision-language models (VLMs) for image editing may struggle to understand negation, implying that they struggle to provide accurate results. One barrier to achieving human-level intelligence is the lack of a standard collection by which research into negation can be evaluated. This thesis presents the first large-scale dataset, Negative Instruction (NeIn), for studying negation within instruction-based image editing. Our dataset comprises 366,957 quintuplets, i.e., source image, …

Contx: Scene Context Prediction Via Context Bank And Layout Perception, Jingxin Liang, Yangyang Xu, Haorui Song, Yuan Lu, Yuhui Deng, Yiyi Long, Yan Huang, Shengxin Liu, Jianbo Jiao, Shengfeng He

Research Collection School Of Computing and Information Systems

Scene context prediction, which seeks to infer unknown contextual information from isolated object properties, currently faces limitations due to predominant reliance on pixel-wise supervision that overlooks real-world context priors. To address this, we present ContX, a context-prior-driven, coarse-to-fine model. ContX distinctively integrates explicit linguistic-contextual knowledge in two key ways. First, it proposes a linguistic guided context bank, leveraging linguistic-statistical contextual data to guide the rationality of segmentation shapes and foster meaningful inter-class contextual interactions. Second, ContX augments contextual comprehension by correlating layouts with linguistic descriptions, enhancing layout perception through a multi-modal strategy. Comprehensive experiments demonstrate ContX's superiority and versatility, outperforming …

Griffin: Effective Token Alignment For Faster Speculative Decoding, Shijing Hu, Jingyang Li, Xingyu Xie, Zhihui Lu, Kim-Chuan Toh, Pan Zhou

Research Collection School Of Computing and Information Systems

Speculative decoding accelerates inference in large language models (LLMs) by generating multiple draft tokens simultaneously. However, existing methods often struggle with token misalignment between the training and decoding phases, limiting their performance. To address this, we propose GRIFFIN, a novel framework that incorporates a token-alignable training strategy and a token-alignable draft model to mitigate misalignment. The training strategy employs a loss masking mechanism to exclude highly misaligned tokens during training, preventing them from negatively impacting the draft model’s optimization. The token-alignable draft model introduces input tokens to correct inconsistencies in generated features. Experiments on LLaMA, Vicuna, Qwen and Mixtral models …

Large Language Models As End-To-End Combinatorial Optimization Solvers, Xia Jiang, Yaoxin Wu, Minshuo Li, Zhiguang Cao, Yingqian Zhang

Research Collection School Of Computing and Information Systems

Combinatorial optimization (CO) problems, central to decision-making scenarios like logistics and manufacturing, are traditionally solved using problem-specific algorithms requiring significant domain expertise. While large language models (LLMs) have shown promise in automating CO problem solving, existing approaches rely on intermediate steps such as code generation or solver invocation, limiting their generality and accessibility. This paper introduces a novel framework that empowers LLMs to serve as end-to-end CO solvers by directly mapping natural language problem descriptions to solutions. We propose a two-stage training strategy: supervised fine-tuning (SFT) imparts LLMs with solution generation patterns from domain-specific solvers, while a feasibility-and-optimality-aware reinforcement learning …

Agentguard: An Active Threat Discovery System For Package Confusion Using Multi-Agent Collaboration, Wei Ma, Yu Li, Zhi Chen, Ye Liu, Lingxiao Jiang, Qiang Hu, Junyi Tao

Research Collection School Of Computing and Information Systems

The proliferation of open-source software (OSS) has made software supply chains prime targets for attacks like Package Confusion, where adversaries publish malicious packages with names deceptively similar to legitimate ones. Existing detection methods often rely on simple lexical similarity or passive analysis of known package pairs, struggle with high false positive rates (FPR), fail to proactively identify emerging threats, and are vulnerable to adversarial evasion. To overcome these limitations, we introduce AgentGuard, a novel framework for proactive, single-input package confusion detection. AgentGuard employs a multi-agent architecture that autonomously discovers potential confusion targets using fine-tuned word embedding model to hybird semantic …

Attachment To Artificial Intelligence: Development Of The Ai Attachment Scale, Construct Validation, And The Psychological Mechanisms Of Human-Ai Attachment, K Tennakoon Appuhamillage Sandeeshwara Kasturiratna, Andree Hartanto

Research Collection School of Social Sciences

Artificial intelligence (AI) systems are increasingly integrated into daily life, not only as tools but also as social partners that people may turn to for interaction and support. This raises important questions about whether, how, and why individuals form attachment-like bonds with AI, and the psychological implications of such attachments. Across five studies involving 1259 unique participants from Singapore and the U.S., the current work developed and validated the 15-item AI Attachment Scale and investigated the dispositional and motivational factors associated with attachment to AI, as well as its emotional and social outcomes. The AI Attachment Scale displayed strong psychometric …

Navigating Ai-Nature Frictions: Autonomous Vehicle Testing And Nature-Based Constraints, Prerona Das, Orlando Woods, Lily Kong

Research Collection College of Integrative Studies

In cities, the application of Artificial Intelligence (AI) is being directed towards transforming different aspects of urban life. These applications take material form in urban spaces, with autonomous vehicles (AVs) providing a prominent example. AI systems rely on large volumes of data on their surroundings to refine the algorithms and enhance the accuracy of prediction for operational efficiency and safety. However, such algorithmic learning and execution can present challenges when dealing with the unpredictable, complex, and dynamic aspects of urban spaces. Nature is a paradigmatic example of such unpredictability, because natural phenomena usually defy consistent patterns and precise data-based modelling. …

A Bayesian Optimisation With Segmentation Approach To Optimising Liquid Handling Parameters, Estefania Yap, Viet Huynh, Calvin Vong, Peter Vogel, Viv Louzado, Thomas Barnes, Buser Say, Michael Burke, Dana Kulić, Aldeida Aleti

Research outputs 2022 to 2026

The automation of liquid handling has become integral in speeding up pharmaceutical development for faster drug development and more affordable treatments. However, the optimal parameters which define the aspirate and dispense procedures vary between liquids and liquid volumes, limiting transfer accuracy and precision. Even state-of-the-art liquid handling devices offer predefined parameters for only a handful of liquids and volumes, resulting in novel parameter sets being defined via a manual, time-consuming process. In this study, we propose an experimental framework for automating the optimisation of liquid class parameters for arbitrary liquids. Within our framework, we propose an optimisation and segmentation algorithm, …

Iostom: Offline Imitation Learning From Observations Via State Transition Occupancy Matching, Quang Anh Pham, Brahmanage Janaka Chathuranga Thilakarathna, Tien Mai, Akshat Kumar

Research Collection School Of Computing and Information Systems

Offline Learning from Observations (LfO) focuses on enabling agents to imitate expert behavior using datasets that contain only expert state trajectories and separate transition data with suboptimal actions. This setting is both practical and critical in real-world scenarios where direct environment interaction or access to expert action labels is costly, risky, or infeasible. Most existing LfO methods attempt to solve this problem through state or state-action occupancy matching. They typically rely on pretraining a discriminator to differentiate between expert and non-expert states, which could introduce errors and instability—especially when the discriminator is poorly trained. While recent discriminator-free methods have emerged, …