6501 research outputs found
Sort by
EFFICIENT SIGNAL SYNTHESIS FOR DATA AUGMENTATION USING GENERATIVE AI
Get PDFModern signal processing AI applications face increasing demands for diverse training data while operating under computational constraints. State-of-the-art generative models, though effective, often require prohibitive resources, limiting their deployment in real-time or embedded systems. This thesis proposes a computationally efficient framework for synthetic signal generation using a two-stage architecture that combines a Vector Quantized Variational Autoencoder (VQ-VAE) with either a decoder-only transformer or a discrete diffusion model. The VQ-VAE encodes high-dimensional signals into discrete latent tokens, significantly reducing model complexity while enabling symbolic sequence modeling. These discrete representations are then modeled using transformer-based autoregressive models or Score Entropy Discrete Diffusion (SEDD) models. We validate this approach on two datasets: TorchSig for radio-frequency signals and AudioMNIST for spoken digits. Our work introduces the first discrete-diffusion based generative models for both audio and RF data and presents the first transformer-based generative model for RF signals trained entirely in discrete latent space. We also improve and extend an existing discrete-space transformer-based speech synthesis pipeline and perform a comprehensive comparative analysis of these generative models across domains. The results demonstrate that these methods maintain high fidelity, generate diverse and realistic signals, and offer substantial computational advantages. This work establishes a scalable foundation for efficient data augmentation in signal-driven machine learning systems and opens new directions for generative modeling in low-resource environments
CREATIVITY AND CRISIS. CASE STUDIES OF BUREAUCRACY UNDER PRESSURE – DETERMINING FACTORS AFFECTING CREATIVITY
No full textI studied creativity and how it interacts with the crisis environment to devise a way of determining if future situations are amenable to creativity. Through studying creativity theories and crisis responses, I developed a framework and instrument that answers the following questions, semi-quantitatively and qualitatively: 1) Is the future situation or problem amenable to creativity? 2) How much “energy” needs to be expended in solving the problem or situation? 3) Where best can I use creative energy as it relates to the subject matter or process of the situation? I studied situations ranging from personal (pursuit of this PhD), critical but not creative (cardiac arrest), and significant U.S. government bureaucratic responses (Apollo 13, 2010 Haiti Response). I evaluated the cases with the variables of importance and urgency, pressure, and two areas of cognitive dissonance (subject matter, process). These variables added together result in a semi-quantitative value related to the amount of energy required in the situation interaction. The cognitive dissonance variables are related to the amount of cognitive energy needed and are a qualitative measurement of the amenability of the situation to creativity. Implications of this research include forecasting future problems that are difficult to solve and require creativity. The final map of variables in a radar plot can highlight areas of the overall problem that might be approached to decrease the cognitive dissonance and thereby free up energy for use in other parts of the problem and the overall system
ENGINEERED PEPTIDE BIOMATERIALS AND SYNTHETIC TRANSMEMBRANE RECEPTORS FOR NATIVE AND PROGRAMMABLE CELL INTERACTIONS
No full textPeptide-functionalized biomaterials enable cell-material interactions through peptide-induced activation of native and synthetic receptors. This dissertation expands the clinical potential of biomaterials by harnessing native and synthetic receptor-peptide interactions. Key advancements include: (1) developing a novel method to functionalize rigid biomaterials with peptides, (2) designing antimicrobial peptide-modified hydrogels and plastics, (3) functionalizing hydrogels with N-cadherin peptides to improve stem cell manufacturing, and (4) engineering synthetic mammalian transmembrane receptors that initiate non-native intracellular responses to peptide-ligands. Using thiol-norbornene click chemistry, we functionalized rigid plastics and soft hydrogels using therapeutic peptides with spatial and temporal control. While advances in synthetic biology have enabled programming mammalian cells with transmembrane receptors responsive to soluble and cell-laden ligands, the development of receptors that respond to material-bound ligands remained limited. To expand the synthetic biology toolbox, we developed EPDA (Extracellular Peptide-ligand Dimerization Actuator) receptors, which allow programmed cells to communicate with peptide-functionalized biomaterials and reversibly activate intracellular responses. These peptide biomaterials and receptor technologies lay the groundwork for future biomedical advancements, particularly in tissue regeneration
THE IMPACT OF LATINO FACULTY ON LATINO STUDENTS
No full textAbstract This study examined the relationship between Latino faculty representation and the academic experiences, sense of belonging, and personal and professional success of Latino students. I used a qualitative case study research method where data was collected through semi structured interviews with self-identified Latino students enrolled in Rowan University Higher Education Administration Program. Participants were prompted to reflect on their collegiate experiences and inspect the role of Latino faculty representation on their educational journey, and the challenges of maneuvering predominantly White academic spaces. The findings revealed that the presence of Latino faculty had a direct impact on students’ feelings of possibility, motivation, and support. Alternatively, the absence of Latino faculty led to feelings of alienation, cultural dissonance, and sense of belonging deteriorating. The study calls for higher education institutions to be intentional about the recruitment, retention, and support services offered to Latino faculty who serve multiple roles as they support students. Overall, the study aimed to center the voices and experiences of Latino students and reinforce that representation goes beyond just being a symbolic gesture, it has direct implications for shaping student success and belonging
When Words Fail: Why AI-Driven Context Beats Word Error Rate for Safety-Critical Air Traffic Control
No full textAccurate speech-to-text transcription is critical for Air Traffic Control (ATC) to en- sure safety and operational efficiency. However, existing Automatic Speech Recognition (ASR) systems struggle with domain-specific phraseology, high noise levels, and real-time constraints. This work explores domain adaptation strategies of OpenAI’s Whisper ASR model, applying Transfer Learning and Low-Rank Adaptation (LoRA) on ATC benchmark and a Federal Aviation Administration (FAA) corpus. Experimental results show that fine- tuning significantly reduces Word Error Rate (WER) from 55.2% to 6.8%, yet WER alone fails to capture safety-critical transcription errors. To address WER’s limitations, we introduce the Contextual Error Rate (CER), a novel evaluation metric powered by a Large Language Model (LLM) that prioritizes errors with operational impact rather than surface-level differences. Our experiments showed that while Transfer Learning improved domain-specific accuracy, LoRA preserved better gen- eralization to conversational English. This research demonstrates the importance of combining domain adapted ASR models with context-aware evaluation metrics. By integrating LLM-assisted CER scoring into the evaluation process, we offer a more reliable and context aware method for evaluating tran- scription quality of domain adapted ASR models in aviation, paving the way for scalable, automated speech recognition in air traffic management. The approach proposed is extensible to other context-critical communication systems where transcription fidelity is paramount. These contributions advance the intersection of speech recognition, machine learning, and aviation safety
Mapping the Path to the Presidency: A Case Study Analysis of New Jersey Four Year Institutions
No full textThe purpose of this research study was to investigate the career trajectories of university presidents at New Jersey’s four-year public and private institutions. Utilizing a qualitative case study design, presidential biographies and CVs were analyzed to identify the professional patterns and pathways leading to the presidential role. Findings revealed that there is a predominance of traditional academic pathways, an increase in non-traditional routes, and distinct differences between the public and private institutions. This study offers implications for leadership development, hiring practices for the senior leadership roles, and for future research on the development of careers leading to senior leadership and the presidency in higher education
Google NotebookLM: What you need to know
No full textLearning objectives: What is Google Notebook LM and how to access Rowan’s licensed version of it Learn different ways to use Google Notebook LM
Google Notebook LM is an AI tool that is licensed through Rowan University. Google NoteBook LM can help with creating study guides for courses and engage with up to 50 different sources at once. Attend this workshop to learn different ways that Google Notebook LM can be used to help you manage information overload. Ethical considerations will be discussed
Artificial Intelligence Policies in Bioethics and Health Humanities: A Comparative Analysis of Publishers and Journals
No full textINTRODUCTION: Rapid advancements in artificial intelligence (AI) pose novel ethical and practical challenges for scholarly publishing. Although AI-related policies are emerging in many disciplines, little is known about the extent and clarity of AI guidance in bioethics and health humanities journals.
METHODS: A search of publicly available journal lists from the American Society for Bioethics and Humanities, Health Humanities Consortium, and Association for Medical Humanities was supplemented with Google Scholar\u27s top 20 bioethics journals ranked by h5-index. This yielded 54 unique journals, of which 50 remained after excluding those without a functional website or recent publications. AI policies were reviewed at the journal and publisher levels were assessed via website review, and editors were contacted for clarification when required. Data extraction was conducted by one author and independently verified by two additional researchers to ensure accuracy.
RESULTS: Of the 50 journals analyzed, only 8 (16%) had a clear AI policy, while 27 (54%) were published by a publisher with an identifiable AI policy. Publisher AI policy statements were favorable to considering AI-assisted manuscripts. Five (10%) of the 8 journals with a clear AI policy explicitly prohibited AI-generated text in submissions. The remaining 15 (30%) journals did not have a publicly available AI policy. Ten of these 15 journals confirmed an absence of any formal AI policy, and seven indicated that discussion to develop guidelines was ongoing.
CONCLUSION: The adoption of AI policies in bioethics and health humanities journals is currently inconsistent. Some journals explicitly ban AI-generated text, whereas others permit AI-assisted writing, with publisher policies being favorable to considering AI-assisted manuscripts. The lack of standardized AI guidelines underscores the need for further discussion to ensure the ethical and responsible integration of AI in academic publishing
Introduction to Rowan Approved AI Tools: Copilot, Adobe, and Gemini
No full textZoom password: d?8JY8iE
This workshop will cover three Rowan AI-approved generators including: Microsoft Copilot, Adobe Firefly, and Google Gemini
Faculty: Mercedes Byrd, Problem-based Learning Librarian and Dr. James Martin II, Assistant Professor, Molecular Biology, Rowan-Virtua School of Osteopathic Medicine
Microsoft Copilot:
Discover Microsoft Copilot\u27s academic applications in this brief hands-on introduction. We\u27ll demonstrate accessing Copilot through our institutional Office 365 environment with a focused example of using it to analyze research data in Excel and generate summary insights. Participants will learn about additional capabilities including document drafting in Word, presentation creation in PowerPoint, and email management. The workshop addresses when Copilot is most effective and concludes with clear guidance on documenting AI assistance in academic work per institutional requirements. By the end, attendees will understand how to integrate Copilot into their workflow while maintaining proper attribution practices.
Adobe Firefly:
Adobe Firefly enables AI-powered visual creation for academic purposes. This quick-start workshop (15 mins) shows participants how to access Firefly through our institutional Adobe Creative Cloud and demonstrates creating a custom infographic for a research presentation. We\u27ll explore additional applications including concept diagrams, presentation graphics, and educational visuals. The session emphasizes understanding when AI-generated images are appropriate for academic use and covers essential attribution requirements for Firefly-created content in publications and presentations. Participants will leave confident in accessing the tool, understanding its academic applications, and properly crediting AI-generated visuals in their scholarly work.
Google Gemini:
This 15-minute introduction will help participants access Google\u27s Gemini AI through our institutional account and understand its role in academic work. We\u27ll demonstrate a live example of using Gemini to synthesize research literature on a specific topic, showing how to craft effective prompts and evaluate responses. Participants will see examples of additional use cases including research question generation, data interpretation, and writing assistance. The session concludes with essential guidance on citing AI assistance according to institutional policies, including proper disclosure in academic papers and presentations. Attendees will leave knowing how to access Gemini, when it\u27s most useful, and how to acknowledge its use appropriately
Julian Marshall Interview
No full textThis conversational inquiry explores the ontology of creativity through an extended dialogue between two educators at the Institute of Contemporary Music Performance. Moving beyond conventional pedagogical frameworks, the conversation grapples with fundamental questions: What is the relationship between technique and creative emergence? How does the teacher create conditions for creativity without systematizing or conceptualizing the creative process itself? What role do trust, validation, and vulnerability play in opening students to their own creative potential? Drawing on diverse philosophical and spiritual traditions--from Heidegger’s ontological inquiry to Buddhist concepts of not-knowing, from Grotowski’s via negativa to the Kabbalistic notion of tzimtzum--Julian Marshall reflects on his forty-year journey as composer, songwriter, and teacher. Through the metaphor of “sliding up” to creativity rather than approaching it head-on, Marshall articulates a pedagogy that privileges presence over methodology, questioning over certainty, and becoming over being. Central to this inquiry is the paradoxical relationship between technical mastery and creative freedom. Marshall argues that creativity happens “through technique” rather than after or beyond it--that rigorous engagement with craft creates the conditions through which something other than the predictable can emerge. Yet he remains wary of reducing this process to a teachable system, insisting that creativity retains an irreducible quality of mystery, grace, and “getting it-ness” that cannot be manufactured or guaranteed. The conversation itself enacts the ontological inquiry it describes: circular rather than linear, associative rather than logical, grappling rather than claiming. As the editor notes, “we circle into truth through stories.” This dialogue offers educators and creative practitioners not a methodology but an invitation--to remain present to the fragile, elusive nature of creativity while honoring the technical foundations that make creative emergence possible