Computer Sciences Commons^™

The Lorekeeper’S Trial: Teaching Ai Literacy Through Active Learning In The Library Classroom, Taylor Greene, Douglas R. Dechow

Library Articles and Research

How can librarians engage students in critical, hands-on learning about artificial intelligence within the limitations of a one-shot session? At Chapman University, librarians have developed an AI literacy session that integrates ethics and hands-on exploration into workshops and course-embedded sessions. This presentation highlights how to weave AI literacy into information literacy instruction, with a focus on a First-Year Foundations program.

Presenters will discuss their efforts to reach students, staff, and faculty through AI literacy initiatives across campus. They will also demonstrate how the Lorekeeper’s Trial—a research quest inspired by RPGs—transforms AI and information literacy concepts into collaborative challenges. Through a …

Does Patient History Influence Capsular Contracture? An Exploratory Analysis With Machine Learning, Thomas M. Johnstone, Daniel Najafali, Jennifer K. Shaw, Justin M. Camacho, Chancellor Johnstone, Rahim S. Nazerali, Gordon K. Lee

Faculty Publications

Background: Capsular contracture (CC) is a frequent and distressing complication of breast augmentation and reconstruction. Although numerous patient-, surgical-, and implant-related risk factors have been proposed, reliable population-level predictors remain inconsistent across studies. This study evaluates whether administrative medical history, as encoded by ICD and CPT codes, contains sufficient predictive signal to identify patients at risk for CC using machine learning. Methods: Patients were queried from the MerativeTM MarketScan® Research Databases from 2003 to 2017 with CPT codes for implant-based breast reconstruction and augmentation. ICD codes were then used to identify all events and conditions of a patient’s history. Hyperparameter-tuned …

Three-Dimensional Shape Cues Affect Human And Artificial Recognition Systems Differently, Mikayla Cutler, Luke D. Baumel, Joseph Tocco, William Friebel, George K. Thiruvathukal, Nicholas Baker Dr.

Computer Science: Faculty Publications and Other Works

Humans and neural networks use shape and texture information differently. While humans weigh shape heavily in their ultimate classification decision, neural networks are more biased towards texture cues. Many tests of shape vs. texture bias have focused on shape recognition from an object’s external contour. However, shape information is also conveyed through internal contours, shading, and attached shadows, especially when an object is viewed from noncanonical perspectives. Using models from ShapeNet, we created datasets of 120,000 texture-substituted images of objects from many viewpoints with and without shading and attached shadows. We tested humans’ and several neural networks’ ability to classify …

A Symbolic Model Of Proof Acquisition In Act-R, Beckett Morris, Kerstin Haring

DU Undergraduate Research Journal Archive

Learning to construct mathematical proofs—formal arguments demonstrating the truth of a mathematical statement using logical deductions and previously established facts—is one of the most challenging skills in STEM education. This research aims to build the foundations for a symbolic cognitive model, using the ACT-R cognitive architecture and implementing in Python with the pyactr package, to explore how different proof strategies can be thought through with only symbols and rules. The model observes simple proofs, and its abilities are assessed based on its generalization capabilities, efficiency, and error patterns. By developing and analyzing such a model, this research provides new insights …

Generation Z And The Ai Misinformation Paradox: Understanding A New Digital Vulnerability, Cecilia Cooley, Elizabeth Sperber

DU Undergraduate Research Journal Archive

This paper asks: How and why is Generation Z more vulnerable to AI-generated misinformation and disinformation than older generations? Using a comparative review of recent empirical studies, survey data, and meta-analyses from 2019–2025, this paper synthesizes research on Gen Z’s exposure to and interaction with AI-produced content across social media platforms. Although it is commonly assumed that Gen Z ’s technological exposure and fluency make them better equipped to recognize false information, findings show the opposite: Gen Z is consistently outperformed by older cohorts in detecting AI-generated falsehoods. This vulnerability stems from three intersecting factors: (1) the sheer volume of …

Du Undergraduate Showcase Abstracts: Research, Scholarship, And Creative Works, Sophia Wismar, Henry Staats, Allison Metzler, Chloe Puckett, Rachel Levine, Christa Kilpatrick, Scott Wolf, Joe Walsh, Grace Doolittle, John Engebreston, Zoe Lopez, Christopher Aaby, Audrey Duff, Timothy Sisk, Katelyn Lamberton, Angela Narayan, Gilkah Argueta, Habiba Samir, Girena Tesfazghi, Genet Kenore, Sinit Tesfamariam, Effley Brooks, Abi Newell, Megan Doherty, Natalie Baer, Lexi Blood, Talya Riciputi, Jessica Jimenez, Devin Hernandez, Lynn Clark, Taj Kumar, Sunil Kumar, Allen Rutman, Mira Pronobis, Tess Carson, Anna Sher, Frankie Stroud, Tamra Pearson D'Estree, Alyssa Wilson, Emily Melnick, Jenalee Doom, Yihang Gao, Gwendolyn Geiger, Noah Gettle, Scott Nichols, Clare Ayoub, Cara Dienno, Sunny Walker, Zoe Hansen, Maya Wheeler, Addison Rice, Patrick Martin, Sanjana Acharya, Daniel Mcintosh, Amanda Mckellips, Calli Cain, Justin Blake, Peter Sokol-Hessner, Natalie Miller, Max Weisbuch, Sophia Dellota, John Macikas, Charlotte Snow, Mark Siemens, Zoe Lynch, Alex Huffman, Prachi Shah, Jason Roney, Halcyon Levi, Nicole Herzog, Andrea Koly, Daniel Linseman, Annie London, Xi Yang, Avery Zwisler, Jane Smith, Chaz Contag, Michael Kerwin, Lucy Rand, Grace Schroeder, Michelle Rozenman, Nissa Tapper, Guiming Zhang, Mateo Mazariego-Halpern, Keith Meyer, Julie Do, Dakota Park-Ozee, Travis Herink, Kara Neu, Jonathan Plomin, Eve-Odine Duchaufour, Debbie Gale Mitchell, Tennyson Anderson-Stricklin, Lily Treitz, Samantha Rosenberger, Sierra Griffith, Finley Joseph, Daniel Sampson, Emmy Davis, Skyler Kasnoff, Evon Lopez, Vivian Nguyen, Cassy Young, Franklin Sellner, Martin Tobon, Ila Graham, Zach Billings, Holden Hedit, Decatur Boland, Paul Kosempel, Cory Chandler, Jay Mahoney, Sam Dragan, Susan Dagget, Yarrow Ator, Heidi Vuletich, Owen Weber, Andrew Kloeppel, Petersen Gray, Mandi Schaeffer-Fry, Razleen Bassra, Bryanna Rodriguez, Christina Blue, Taubie Sanders, Rachel Epstein, Luke Milburn, Camryn Evans, Ezra Martinez, Mary Westwood, Gabri Notov, Robin Tinghitella, Lilou Cabrol, Eli Barbour, Juliet Mendik, Selma Myers, Zac Wise, Noah Fahlin, Michelle Knowles, Abigail Hopper, Michael Greenberger, Romi Laclair, Sarah Watamura, Sabrina Efroymson, Casey Barker, Sydney Seltzer, Bryn Yehle, Jennifer Hoffman, Sara Garcia, Ryuka Nagamine, Trevor Briggs, Remy Le Boeuf, Elena Krone, Eileen Farrell, Regan O'Rourke, Elena Roel, Greg Mortimer, Ali Ayoub, Stefani Langehennig, Caitlin Turk, Logan Scmid, Stefan Chavez-Norgaard, Karen Kim, Tatiana Peccedi, Courtney Cassidy, John Sebesta, Rhianna Lewis, Janice Bening-Lacek, Vivian Lawless, Mckenna Hanson, Jeffrey Amidon, Riya Joshi, Ram Ambre, Brady Worrell, Perrin Schneider, Ali Azadani, Brooke Agulnek, Lyndsie Salvagio, Elise Siemanowki, Yan Qin, Andre Allen, Melodie Nguyen, Megan Livengood, Abby Reams, Saffron Hartreeve, Bri Wylie, Sarah Brookman, Mariah Loiacono, Green Russo, Abhia Lodhi, Gabrielle Welsh, Nika Spehar, Shahked Levin, Evrim Baykal, Kimberly Chiew, Jocelyn Torres, Kailey Hicks, Mykaela Tanino-Springsteen, Audrey Bellows, Akam Chahal, Madeline Tepper, Shannon Murphy, Alexa Fonseca, Deborah Han, Cassandra Perez, Oluwatoyin Alaba, Julia Roncoroni, Vy Nguyen, Nana Burn, Sarah Sasse, Rubin Tuder, Anthony Gerber, Nancy Lorenzon, Christine Vohwinkel, Camryn Gunter, Tristan Weber, Sam Rommel, Brian Michel, Muskan Fatima, Alannah Oleson, Kira Frey, Edward Garrido, Beckett Morris, Kerstin Haring, Drew Middleton, Abigail Walpert, Liam Dee, Gabby Ishaw, Cole Carnes, Maddie Weiser, Claire Fox, Valeriia Vlasenko, Kateri Mcrae, Riley Smith, Abigail Templin, Kushani Rajapaksha

DU Undergraduate Research Journal Archive

Abstracts from the DU Undergraduate Research Showcase.

The 1st International Workshop On Foundations And Architectures For The Agentic Web, Boualem Benatallah, Pradyumna Chari, Abul Ehtesham, Abderrahmane Maaradji, Luca Muscariello, Fatma Outay, Ramesh Raskar, Yacine Sam, Sabrina Senatore, Aditi Singh

All Works

The Agentic Web is emerging as billions of AI agents discover, communicate, and coordinate across the open Web, shifting from isolated models to Web-Inetgrated entities. This workshop explores the state of the art, open challenges, and emerging research directions in the foundations and architectures of the Agentic Web. It provides a forum for researchers and practitioners to examine interoperable architectures, protocols, and standards enabling AI agents to operate as first-class Web entities. Beyond technical interoperability, it also adresses economic and societal mechanisms including reputation, governance, accountability, and large-scale coordination.

Panda-Plus-Bench: A Clinical Benchmark For Evaluating The Robustness Of Ai Foundation Models In Prostate Cancer Diagnosis, Joshua L. Ebbert, Dennis Della Corte

Faculty Publications

Artificial intelligence foundation models are increasingly deployed for prostate cancer Gleason grading, where GP3/GP4 distinction directly impacts treatment decisions (active surveillance vs. intervention). However, these models may achieve high validation accuracy by learning specimen-specific artifacts rather than generalizable biological features, limiting real-world clinical utility. We introduce PANDA-PLUS-Bench, a curated benchmark dataset derived from expertly annotated prostate biopsies designed specifically to quantify this failure mode. The benchmark comprises nine carefully selected whole slide images from nine unique patients containing diverse Gleason patterns, with non-overlapping tissue patches extracted at both 512 × 512 and 224 × 224-pixel resolutions across eight augmentation conditions. …

Optimal-Cost Construction Of Shallow Cuttings For 3-D Dominance Ranges In The I/O-Model, Yakov Nekrich, Saladi Rahul

Michigan Tech Publications

Shallow cuttings are a fundamental tool in computational geometry and spatial databases for solving offline and online range searching problems. For a set P of N points in 3-D, at SODA’14, Afshani and Tsakalidis designed an optimal O(N log2 N) time algorithm that constructs shallow cuttings for 3-D dominance ranges in internal memory. Even though shallow cuttings are used in the I/O-model to design space and query efficient range searching data structures, an efficient construction of them is not known till now. In this paper, we design an optimal-cost algorithm to construct shallow cuttings for 3-D dominance ranges. The number …

High-Throughput Robotic Ethanol Inhibition Assays For Engineered Thermophilic Biofuel Strains, Kevin He, Daniel Olson, Marybeth Maloney, Anthony Lanahan

Wetterhahn Science Symposium Posters

Ethanol stress assays are commonly used to evaluate microbial tolerance, metabolic adaptation, and fermentation performance. However, manual liquid handling introduces variability across replicate wells and small-volume pipetting steps, limiting reproducibility and throughput. This study developed an automated OT-2 robotic workflow to generate replicated ethanol concentration gradients for high-throughput inhibition assays in engineered thermophilic biofuel strains. Kinetic plate-reader measurements were used to quantify ethanol-dependent growth responses under anaerobic fermentation conditions. The reasearch question is: How do engineered thermophilic biofuel strains differ in ethanol-dependent growth inhibition under anaerobic fermentation conditions, and can automated robotic assays improve the reproducibility of these measurements? Can …

Distributed File Carving, Brad J. Baudin

LSU Master's Theses

File carving is a fundamental technique in the digital forensics community, enabling analysts to recover deleted files from disk images without relying on filesystem metadata; however, as storage capacities continue to increase, modern file carving tools face significant scalability challenges, with carving time growing substantially alongside disk image size, particularly in the presence of file fragmentation. Fragmentation, a common behavior in modern filesystems, distributes file data across non-contiguous disk blocks to maximize space utilization, and in large disk images this distribution can span wide logical distances, increasing the search space and reducing carving efficiency, causing traditional approaches to struggle within …

Ai-Infused Writing-Intensive Problems For Introduction To Programming In C++, Esma Yildirim

Open Educational Resources

This exercise book uses Writing-Intensive (WI) pedagogy in an introductory programming course setting, because I believe WI can help students engage in deeper analysis of programming problems while strengthening their critical and analytical thinking skills. In computer science, a single problem can often be solved using multiple algorithms, designs, and implementation strategies. Exploratory thinking and reflective writing can improve students’ ability not only to read and write computer programs, but also to evaluate the efficiency, readability, and maintainability of algorithms and code. Students must learn not only how to design solutions, but also how to compare alternatives and determine the …

Enhanced Population Mean Estimation Using An Exponential Estimator With Known Medians Of Dual Auxiliary Variables Within A Neutrosophic Approach: Applications In Agricultural Yield Prediction, Anchal Yadav, Mukesh Kumar

Neutrosophic Systems with Applications

In classical statistical theory, estimation of population parameters is generally carried out under the assumption that all observed data are precise, complete, and free from ambiguity. However, in many practical and real-world situations, data often deviate from these ideal conditions and instead appear in vague, uncertain, or interval-valued forms. Such imperfections reduce the effectiveness of traditional estimation techniques and motivate the development of more flexible and robust methodologies. To address these challenges, several improved estimators, particularly neutrosophic ratio-type estimators and their advanced extensions, have been proposed in recent literature. In this study, a new estimator known as the two auxiliary …

On Some Properties Of Generalised Regular Anti-Topological Space, Bhimraj Basumatary, Gugu Narzary, Jeevan Krishna Khaklary

Neutrosophic Systems with Applications

This article studies Regular Anti-open sets and Generalised Regular Anti-open sets within a topological framework, addressing the limited development of anti-open structures in generalised topology. We introduce Regular Anti-open sets and establish their fundamental properties. The study is extended by defining Generalised Regular Anti-open sets, providing a broader and more flexible class of sets. Furthermore, the notions of GR-interior and GR-closure are introduced and analyzed, and their essential properties are obtained. The results contribute to a clearer understanding of generalized anti-open structures and provide a basis for further research in topology.

Optimization Of A Multi-Item Supply Chain Model With Shortages Under Parametric Type-2 Interval Environments, Sourav Kumar Kumar Giri, Totan Garai, Sahidul Islam, Haridas Mondal, Shariful Alam

Neutrosophic Systems with Applications

Modern supply chain systems frequently operate in environments where demand, costs and inventory-related parameters are uncertain and difficult to estimate accurately. These uncertainties become more critical in multi-objective decision-making situations, where decision makers must simultaneously balance several conflicting goals. Conventional optimization techniques often fail to represent the ambiguity and vagueness present in practical decision environments. To overcome these limitations, this study develops a multi-item supply chain model for a single supplier and a single buyer by incorporating Type-2 interval representations into the modelling framework. The proposed approach introduces a structured set of arithmetic operations for Type-2 intervals to manage uncertain …

Classes Of Analytic Functions Defined By Salagean Derivative Operator Associated With Neutrosophic Generalized Poisson Distribution, Soliu O. Opeyemi Okunola, Olushola Adeyemo, Sayo A. Abidemi Gbangbala, Folorunso I. Isola Akinwale

Neutrosophic Systems with Applications

This study introduces and analyses new subclasses of analytic functions by applying the Salagean derivative operator to the Neutrosophic Generalized Poisson Distribution (NGPD) series. We develop a model where the mean parameter is treated as an interval or set to account for indeterminacy in complex systems. By employing Stirling numbers of the second kind and decreasing factorials, we derive necessary and sufficient coefficient inequalities and inclusion relations for these new subclasses. Numerical results and graphical illustrations demonstrate the sensitivity of these functions to orientation and the neutrosophic parameter, providing a framework for applications in fields like medical imaging and network …

A Goal Programming Approach For Finding The Best Compromise Solution Of Multiobjective Linear Programming Under Neutrosophic Environment, Sultan S. Alodhaibi, Hamiden Abd El- Wahed Khalifa

Neutrosophic Systems with Applications

Decision problems often involve multiple objectives that conflict, making simultaneous optimization impossible. These problems require trade-offs to achieve a solution that balances the competing goals. In this paper, the detailed discussion that is related to linear programming (SVTrNFMOLP) with single valued trapezoidal neutrosophic numbers is made. Furthermore, this study deals with all the related parameters and discussion. As rank function and due to its definition, the SVTrNFMOLP is transformed in the crisp MOLP. It is noticed that goal programming is best to get the best compromise solution. The advantages of the proposed approach are: The use of Tr allows the …

A Govsecops-Oriented Governance, Risk, And Compliance Platform For Continuous Authorization In Dod Il4/Il5 Environments, Anand Janjal

Journal of Cybersecurity Education, Research and Practice

Department of Defense (DoD) Impact Level 4 and Impact Level 5 (IL4/IL5) systems require continuous assurance of cybersecurity posture under stringent operational and regulatory constraints. While the NIST Risk Management Framework (RMF) and DoD DevSecOps guidance emphasize continuous authorization supported by real-time evidence, many existing Governance, Risk, and Compliance (GRC) platforms remain documentation-centric and insufficiently integrated with operational telemetry. This paper presents a GovSecOps-oriented GRC architecture that integrates vulnerability ingestion, automated Plan of Action and Milestones (POA&M) lifecycle management, dynamic risk scoring, and continuous authorization dashboards within IL4/IL5 environments. Using a Design Science Research methodology, the study develops and evaluates …

Effects Of Coulomb Collisions On The Structures Of Fast Magnetosonic Shocks, Bo Farnell

Physics and Astronomy Undergraduate Senior Theses

We present fully kinetic particle-in-cell simulations demonstrating qualitative effects of binary Coulomb collisions on fast magnetosonic shocks. We find that with a sufficiently high collisional frequency, shock rippling and reformation can be inhibited, creating stationary, laminar shocks. We see that collisions rapidly bring the reflected population into equilibrium with the upstream population, adding credibility to existing theories that the separation of these two populations create these dynamical behaviors around the shock transition region. We also see that Ohmic heating from Coulomb collisions can suppress instabilities.

Epileptic Seizure Prediction From Eeg Using Continual Learning With Cnns, Adnan Amin, Ammar Bathich, Feras Al-Obeidat, Safa Naes, Maria Jose Sousa

All Works

Epilepsy is a persistent neurological disorder that affects over 50 million people worldwide, with nearly one-third of patients remaining unresponsive to conventional therapeutic treatments. This study introduces a progressively adaptive seizure prediction framework designed to enhance early detection and clinical decision-making. The proposed model employs a deep learning strategy grounded in continual learning (CL) principles, using Convolutional Neural Networks (CNNs) in combination with knowledge distillation techniques. This enables the model to assimilate new data while retaining previously learned information. The approach was evaluated on the publicly available Bonn University EEG dataset, following a sequential learning process in which each successive …

Vigor: Virtual Intelligence For Gaze Observation And Representation, Meherun Nesa Shraboni

Theses and Dissertations

Eye tracking in modern XR head-mounted displays can capture high-frequency gaze data at 90–120 Hz, generating thousands of samples within a single 30-second interaction. At the same time, the XR market has grown to millions of active users worldwide, with major platforms such as Meta investing heavily in eye-tracking-enabled devices. Despite this large-scale adoption and data availability, most applications still rely on hand-controller input, with gaze either processed offline or reduced to a simple pointing signal. As a result, the majority of temporally rich gaze information remains unused in real-time interaction, limiting system responsiveness, reducing interaction fidelity, and preventing effective …

Replication And Architectural Enhancement Of Omni Aggregation Networks, Matteo Calviello

Rose-Hulman Undergraduate Research Publications

Single Image Super-Resolution (SISR) reconstructs high-resolution images from low-resolution inputs, a capability valuable across many application domains. This thesis focuses on the replication and architectural enhancement of the Omni Aggregation Network (OAN), a lightweight super-resolution architecture designed to balance reconstruction quality with computational efficiency through a novel omni-axis self-attention mechanism.

Independent replication remains a persistent challenge in machine learning research, and this work addresses that issue directly. Initial replication attempts trailed the original reported results by approximately 6-10% in performance metrics before several undocumented implementation details were identified. After correcting these discrepancies, the replicated implementation achieved results within 4% of …

Replication And Architectural Enhancement Of Omni Aggregation Networks, Matteo Calviello

Senior Projects - Computer Science & Software Engineering

Single Image Super-Resolution (SISR) reconstructs high-resolution images from low-resolution inputs, a capability valuable across many application domains. This thesis focuses on the replication and architectural enhancement of the Omni Aggregation Network (OAN), a lightweight super-resolution architecture designed to balance reconstruction quality with computational efficiency through a novel omni-axis self-attention mechanism.

Independent replication remains a persistent challenge in machine learning research, and this work addresses that issue directly. Initial replication attempts trailed the original reported results by approximately 6-10% in performance metrics before several undocumented implementation details were identified. After correcting these discrepancies, the replicated implementation achieved results within 4% of …

Finding A Way Out Of The Filter Bubble: The Confusion Of A Heavy Social Media User, Longwen Miao

Masters Theses

This thesis studies how algorithmic recommendation systems reshape visual perception, aesthetic judgment, and the construction of selfhood within contemporary digital culture.

Everything begins with the experience of repeatedly encountering algorithmically recommended content on everyday digital platforms. On these platforms, images, sounds, and social interactions are continuously selected, repeated, and reorganized by predictive systems, forming an environment in which perception is constantly structured and adjusted.

From this observation, the research raises two central questions: how do algorithmic recommendation systems reshape visual perception, aesthetic judgment, and self-recognition, and how might these systems be intervened in or made perceptible through artistic practice? Within …

Frictional Intelligence, Posheng Cheng

Masters Theses

This is an experimental interaction design project that challenges anthropomorphism in human-computer interaction. In particular, the recent advancement of artificial intelligence technologies like Large Language Models has taken anthropomorphism to new heights. The conversational chatbot interface of AI prioritizes mimicking an inherently human communication medium to maximize human-likeness. However, anthropomorphism has several downsides. Conversational interfaces obscure the limitations and the tangible cost of the technology. They also imply fictional moral status and human-level cognitive capabilities, which means general public sentiment focuses on the ``overhyped'' excitement and fear rather than on other socio-ethical and capacity questions that are far more urgent …

Ai-Driven Vehicular Federated Learning: Fairness-Aware Adaptive Incentives With Blockchain Verifiability For Smart Transportation, Abir Raza

Thesis/ Dissertation Defenses

Vehicular Federated Learning (VFL) is becoming a crucial enabler for the implementation and optimization of automated transport systems. This technology enables intelligent transportation systems to function by enabling networked vehicles to develop perception and control models through joint training while preserving their original data. Nevertheless, the effective deployment of VFL is dependent on the sustained participation of trustworthy vehicles. However, the dynamic nature of vehicular environments poses critical challenges, including unstable participation, data heterogeneity, and resource constraints. The sustained collaboration of smart vehicles necessitates reliable client selection and equitable incentive schemes with verifiable transparency. The distributed nature of VFL makes …

Hierarchical Motion Planning Of Mobile Robot Based On Dynamic Corridor Inflation And Convex Optimization, Dingkun Zhang, Haizhao Liang

Journal of System Simulation

Motion planning for robots with Ackermann chassis in dynamic complex environments faces nonholonomic constraints and kinematic-dynamic coupling challenges. However, traditional methods suffer from path redundancy, random fluctuations, and local optimality. A hierarchical motion planning method based on dynamic corridor inflation and convex optimization is proposed. Topologically sparse paths are generated by fusing the Ramer-Douglas-Peucker (RDP) path compression operator with the A* algorithm to reduce redundant path points' interference with backend optimization. Dynamic corridor inflation strategies are designed considering Ackermann steering characteristics, and safe corridors satisfying kinematic constraints are constructed via convex decomposition. Corridor constraints are then transformed into linear inequalities …

Detection Method For Laboratory Ppe Compliance Wearing Based On Human Key Points, Lijun Peng, Tingqi Su, Peijin Liu, Lin He, Xiewu Zhou, Minxin Zhang

Journal of System Simulation

To address the problems of high missed detection rate and inaccurate judgment of wearing compliance when multi-scale and multi-category targets of laboratory personnel's safety protective equipment are detected in a complex laboratory environment, this paper proposes a laboratory personnel's standard personal protective equipment (PPE) wearing detection method (multi-scale multi- target joint key point detection method, MSMT-JKDM) that integrates multi-scale features and human keypoints. The multi-scale adaptive down sampling (MSA-Down) module and the cascaded group attention transformer (CGA Former) are introduced to enhance the feature representation ability of PPE (especially small targets such as goggles and gloves) in laboratory detection scenarios, …

Optimizing Gated Rnns, Joshua Paul Fechete

Honors Projects

Gated recurrent neural networks such as Long Short-Term Memory (LSTM) and Gated Recurrent Units (GRU) help fix instability present in normal recurrent neural networks. This allows them to be used for various real-world tasks, and due to their architecture, they are uniquely qualified to handle variable sized input such as text. However, even before training can begin on a machine learning model, various hyperparameters must be chosen to decide how the model will be architectured. Choosing good hyperparameters is vital for creating a model that performs well but is not larger and more computationally expensive to run than it needs …

Power Flow Calculation Based On Block-Encoded Adiabatic Quantum Newton-Raphson Method, Shengchao Jiang, Yunqing Pei, Hongying Zhai, Guojian Wu, Fang Gao

Journal of System Simulation

To overcome the efficiency bottleneck of the traditional Newton-Raphson (NR)method in high- dimensional power flow calculations for modern power systems and the constraints of variational quantum algorithm frameworks, this paper proposed a power flow calculation framework integrating block encoding technology and adiabatic quantum computing principles. Based on block encoding technology, adiabatic quantum theory, and the NR method, a block-encoded adiabatic quantum power flow calculation framework (BQ-NR) was constructed. The NR correction equations were mapped to a quantum system, and the quantum state encoding of the correction equations was realized by constructing an extended Hermitian matrix and a projection operator; a …