Artificial Intelligence and Robotics Commons^™

Hop‑Based Heterogeneous Graph Transformer, Zixuan Yang, Xiao Wang, Yanhua Yu, Yuling Wang, Kangkang Lu, Zirui Guo, Xiting Qin, Yunshan Ma, Tat‑Seng Chua

Research Collection School Of Computing and Information Systems

The Graph Transformer (GT) has shown significant ability in processing graph-structured data, addressing limitations in graph neural networks, such as over-smoothing and over-squashing. However, the implementation of GT in real-world heterogeneous graphs (HGs) with complex topology continues to present numerous challenges. Firstly, a challenge arises in designing a tokenizer that is compatible with heterogeneity. Secondly, the complexity of the transformer hampers the acquisition of high-order neighbor information in HGs. In this paper, we propose a novel Hop-basedHeterogeneous Graph Transformer (H2Gormer) framework, paving a promising path for HGs to benefit from the capabilities of Transformers. We propose a Heterogeneous Hop-based Token …

Performing Information Extraction For Mission Engineering Applications, Samuel R. Koski

Engineering Management & Systems Engineering Theses & Dissertations

The process of extracting structured data from unstructured and semi-structured text is manual, time consuming and error prone. Current natural language processing approaches for automating this process are difficult to verify for non-trivial and context-sensitive corpora. Large Language Models (LLMs) like ChatGPT have become a subject of considerable interest, opening a promising avenue of exploration. However, there is limited evidence on the performance of LLMs for information extraction.

In this dissertation, an approach is proposed to evaluate the accuracy of Stanford OpenIE and OpenAI's ChatGPT for this purpose. This includes comparing Resource Description Framework (RDF) triples extracted by each of …

Computational Modeling And Analysis Of Facial Expressions And Gaze For Discovery Of Candidate Behavioral Biomarkers For Children And Young Adults With Autism Spectrum Disorder, Megan Anita Witherow

Electrical & Computer Engineering Theses & Dissertations

Facial expression production and perception in autism spectrum disorder (ASD) suggest the potential presence of behavioral biomarkers that may stratify individuals on the spectrum into prognostic or treatment subgroups. High-speed internet and the ease of technology have enabled remote, scalable, affordable, and timely access to medical care, such as measurements of ASDrelated behaviors in familiar environments to complement clinical observation. Machine and deep learning (DL)-based analysis of video tracking (VT) of expression production and eye tracking (ET) of expression perception may aid stratification biomarker discovery for children and young adults with ASD. However, there are open challenges in 1) facial …

Time Series Models For Predicting Application Gpu Utilization And Power Draw Based On Trace Data, Dorothy Xiaoshuang Parry

Electrical & Computer Engineering Theses & Dissertations

This work explores collecting performance metrics and leveraging various statistical and machine learning time series predictive models on a memory-intensive application, Inception v3. Trace data collected using nvidia-smi measured GPU utilization and power draw for two runs of Inception3. Experimental results from the statistical and machine learning-based time series predictive algorithms showed that the predictions from statistical-based models were unable to capture the complex changes in the trace data. The Probabilistic TNN model provided the best results for the power draw trace, according to the test evaluation metrics. For the GPU utilization trace, the RNN models produced the most accurate …

Scaled And Graduated Learning In Deep Relu Networks And Reconstructing Depp Inelastic Scattering Kinematics, Abdullah Ayar Farhat

Mathematics & Statistics Theses & Dissertations

To address computational challenges in learning deep neural networks, properties of deep RELU networks were studied to develop a multi-scale learning model. The multi-scale model was compared to the multi-grade learning models. Unlike the deep neural network learned from the standard single-scale, single-grade model, the multi-scale neural networks use low scale information from all hidden layers, and thusly provide a robust approximation method that requires fewer parameters, lower computational time, and is resistant to noise. It is shown that the multiscale method is not subject to issues arising from the vanishing gradient problem. This allows very deep multi-scale networks to …

Multi-Modality Transformer For E-Commerce: Inferring User Purchase Intention To Bridge The Query-Product Gap, Srivatsa Mallapragada

Dissertations

The rapid growth of e-commerce has necessitated the development of sophisticated product retrieval systems that can effectively match user queries with relevant products. However, the semantic gap between queries and products remains a significant challenge, as traditional retrieval methods often fail to capture the nuances of user purchase intentions. E-commerce click-stream data and product catalogs offer critical user behavior insights and product knowledge that are untapped in the current product search algorithms. This dissertation presents learning strategies that leverage the query-product transaction logs to enrich the pipeline of our proposed multi-modal transformer model, which transforms initial user queries into pseudo …

Ai And Design Thinking: The Human Connection, Ellen Ramsey

Faculty and Staff Publications & Presentations

No abstract provided.

A Gateway To Next-Generation Patient Monitoring System, Kishore Kumar Kadari

USF Tampa Graduate Theses and Dissertations

Healthcare patient monitoring is undergoing a significant digital transformation, and the integration of Cyber-Physical Systems (CPS) and Artificial Intelligence (AI) is becoming increasingly crucial in reshaping patient care. In an era where digital technology is revolutionizing medical practices, this research aims to take a leading role in advancing a fundamental aspect of predictive and sustainable healthcare practices, enhancing patient outcomes and uplifting the practice of medicine.

This research focuses on the study of Digital Twins for precision health, which are designed to monitor and provide intricate, personalized feedback dynamically during a patient's healthcare experience. The architecture of the system is …

Dyvir: Virtual Reality Generated Synthetic Training Datasets For Ai, Garrett Williams

Graduate Student and Postdoctoral Fellow Symposium

Artificial Intelligence (AI) can perform complex tasks quickly such as object detection. To perform these tasks, the AI algorithms are first trained on data. However, some data such as labeled imagery of aerial objects is hard to obtain. Utilizing Virtual Reality (VR) software, a custom tool called DyViR was made to generate synthetic training datasets. Users customize the virtual environment, aerial objects, and sensor modality to produce custom-tailored datasets.

Preserving Linguistic Diversity In The Digital Age: A Scalable Model For Cultural Heritage Continuity, James Hutson, Pace Ellsworth, Matt Ellsworth

Faculty Scholarship

In the face of the rapid erosion of both tangible and intangible cultural heritage globally, the urgency for effective, wide-ranging preservation methods has never been greater. Traditional approaches in cultural preservation often focus narrowly on specific niches, overlooking the broader cultural tapestry, particularly the preservation of everyday cultural elements. This article addresses this critical gap by advocating for a comprehensive, scalable model for cultural preservation that leverages machine learning and big data analytics. This model aims to document and archive a diverse range of cultural artifacts, encompassing both extraordinary and mundane aspects of heritage. A central issue highlighted in the …

An Automated Approach For Improving The Inference Latency And Energy Efficiency Of Pretrained Cnns By Removing Irrelevant Pixels With Focused Convolutions, Caleb Tung, Nick Eliopoulos, Purvish Jajal, Gowri Ramshankar, Chen-Yun Yang, Nicholas Synovic, Xuecen Zhang, Vipin Chaudhary, George K. Thiruvathukal, Yung-Hsiang Lu

Computer Science: Faculty Publications and Other Works

Computer vision often uses highly accurate Convolutional Neural Networks (CNNs), but these deep learning models are associated with ever-increasing energy and computation requirements. Producing more energy-efficient CNNs often requires model training which can be cost-prohibitive. We propose a novel, automated method to make a pretrained CNN more energy-efficient without re-training. Given a pretrained CNN, we insert a threshold layer that filters activations from the preceding layers to identify regions of the image that are irrelevant, i.e. can be ignored by the following layers while maintaining accuracy. Our modified focused convolution operation saves inference latency (by up to 25%) and energy …

Data Supporting Research On Personalized Learning Paths, Sean Mochocki, Mark Reith

Faculty Publications

Personalized Learning Paths (PLPs) are a key application of Artificial Intelligence in E-Learning. In contrast to regular Learning Paths, they return a unique sequence of learning materials identified as meeting the individual needs of the students. In the literature, PLPs are often created from knowledge graphs, which assist with ordering topics and their associated learning materials. Knowledge graphs are typically directed and acyclic, to capture prerequisite relationships between topics, though they can also have bidirectional edges when these prerequisite relationships are not necessary. This data package provides a primarily un-directed knowledge graph, with associated repository of open-source learning materials that …

Icolc Statement On Ai In Licensing, International Coalition Of Library Consortia

Copyright, Fair Use, Scholarly Communication, etc.

The International Coalition of Library Consortia (ICOLC) statement on artifical intelligence in licensing.

Unraveling Biases And Customer Heterogeneity In E-Commerce Recommendation Systems, Sachin Sharma

Dissertations

This research explores the biases present in AI algorithms within e-commerce recommendation systems, focusing on how these biases prioritize popular, sponsored, and private-label products over actual customer preferences. We extend the responsible AI discourse by critically examining these biases and their implications for fairness in e-commerce. To strengthen the current understanding of AI fairness in the fields of information systems and computer science, we aim to challenge the assumption that AI fairness is objective and the same for everyone. We examine how individual differences, such as equity sensitivity and exchange ideology, contribute to users' varied perceptions of AI fairness. Through …

Challenges On Public Security System In Ai Era—Preface For Special Column “Artificial Intelligence And Public Security”, Juan Cao, Qiang Sheng, Guojie Li

Bulletin of Chinese Academy of Sciences (Chinese Version)

The rapid development of artificial intelligence generated content (AIGC) technology has triggered new public security risks, posing a serious threat to national security and social stability. This study exhibits the recent advances of artificial intelligence (AI) content generation and detection techniques, points out the challenges of detection techniques in real-world scenarios, and advocates that it is necessary to develop AIGC detection technology for public security needs and build a whole-process detection technology system from generative models to online platforms, which supports AIGC to be labeled at the generation phase, identifiable during dissemination and source-traceable after the incident occurs.

Editorials For Spencial Topic “Scientific Focus: Open Source Innovation And Open Source Paradigm”

Bulletin of Chinese Academy of Sciences (Chinese Version)

No abstract provided.

Open Source In China: Opportunities In New Era, Qigang Zhu, Guofeng Zhang, Caihua Zhu, Yi Zhang

Bulletin of Chinese Academy of Sciences (Chinese Version)

China has rapidly experienced industrialization and informatization, entering the era of digital economy. The combination of new productive forces represented by artificial intelligence, non-exclusive production factors represented by knowledge, and open source related relationships may form a new model. From the perspective of social practices of open source innovation, open source lacks systematic preparedness at theoretical, institutional, and talent levels. China should seize the historical opportunity to complete theoretical updates and cultural reconstruction those are adapted to it. Open source and openness also have conditions for China to create a global paradigm of cooperation, sharing, and innovation in the digital …

Thoughts On Ai Innovation And Open Source Development: Lessons From Deepseek, Yanjun Wu

Bulletin of Chinese Academy of Sciences (Chinese Version)

At a critical moment of intense competition in the artificial intelligence (AI) field, DeepSeek has released foundational large language models (LLM) such as V3/R1, with performance comparable to leading international organizations like OpenAI. This not only demonstrates China’s technological innovation capabilities in AI but also provides a Chinese innovative pathway for global AI development. Firstly, through low-cost training and inference, break the monopolistic barriers of high-end computing power and lower research and development thresholds. Secondly, through full-stack and comprehensive open-source strategies, support customizable and local deployment that benefits various industries. This technological innovation and open-source practice from DeepSeek deserves in-depth …

Current Situation, Challenges And Policy Suggestions Of Chinese Enterprises Participation In Open Source Innovation, Kaihua Chen, Xin Wen, Qigang Zhu

Bulletin of Chinese Academy of Sciences (Chinese Version)

Open source is emerging as a new paradigm and a crucial driving force for technological innovation and industrial development. It is shaping new impetus for the development of the digital economy, promoting the construction of an innovation ecosystem in the digital era, and it also serves as an effective path for Chinese enterprises to catch up in technology and industry. Through investigations and literature research, this study discovers that Chinese enterprises have achieved remarkable progress in the field of open source, with the scale of participation continuously expanding and the influence steadily increasing. Nevertheless, they still face numerous challenges in …

Using Chatgpt To Generate Gendered Language, Shweta Soundararajan, Manuela Nayantara Jeyaraj, Sarah Jane Delany

Conference papers

Gendered language is the use of words that denote an individual's gender. This can be explicit where the gender is evident in the actual word used, e.g. mother, she, man, but it can also be implicit where social roles or behaviours can signal an individual's gender - for example, expectations that women display communal traits (e.g., affectionate, caring, gentle) and men display agentic traits (e.g., assertive, competitive, decisive). The use of gendered language in NLP systems can perpetuate gender stereotypes and bias. This paper proposes an approach to generating gendered language datasets using ChatGPT which will provide data for data-driven …