Search

Hackathons have emerged as pivotal platforms in the software industry, driving both innovation and skill development for organizations and students alike. These events enable companies to quickly prototype new ideas while offering students practical, hands-on learning experiences. Over time, hackathons have transitioned from purely competitive events to valuable educational tools, integrating theory with real-world problem-solving through collaboration between academia and industry. The infusion of artificial intelligence (AI) and machine learning is now reshaping hackathons, providing enhanced learning opportunities while also introducing ethical challenges. This study explores the influence of generative AI on students' technological choices, focusing on a case study from the 2023 University of Iowa Hackathon. The findings offer insights into AI's role in these events, its educational impact, and propose strategies for integrating such technologies in future hackathons, ensuring a balance between innovation, ethics, and educational value., Comment: 9792 words, 26 pages, 12 figures

Fluidic channels at atomic scales regulate cellular trafficking and molecular filtration across membranes and thus play crucial roles in the functioning of living systems. However, constructing synthetic channels experimentally at these scales has been a significant challenge due to the limitations in nanofabrication techniques and the surface roughness of the commonly used materials. Angstrom-scale slit-like channels address this challenge, as these can be made with precise control over their dimensions and can be used to study the fluidic properties of gases, ions and water at unprecedented scales. Here, we provide a detailed fabrication method of the two-dimensional (2D) angstrom-scale channels, which can be assembled as a single channel or up to hundreds of channels made with atomic scale precision using layered crystals. The procedure includes the fabrication of the substrate, flake, spacer layer, flake transfers, van der Waals assembly, and post-processing. We further explain how to perform molecular transport measurements with the angstrom-scale channels, for the development of methods directed at unravelling interesting and anomalous phenomena that help shed light on the physics of nanofluidic transport systems. The procedure requires a total of 1 to 2 weeks for the fabrication of the 2D channel device and is suitable for users with prior experience in clean room working environments and nanofabrication.

This research study explores the conceptualization, development, and deployment of an innovative learning analytics tool, leveraging OpenAI's GPT-4 model to quantify student engagement, map learning progression, and evaluate diverse instructional strategies within an educational context. By analyzing critical data points such as students' stress levels, curiosity, confusion, agitation, topic preferences, and study methods, the tool provides a comprehensive view of the learning environment. It also employs Bloom's taxonomy to assess cognitive development based on student inquiries. In addition to technical evaluation through synthetic data, feedback from a survey of teaching faculty at the University of Iowa was collected to gauge perceived benefits and challenges. Faculty recognized the tool's potential to enhance instructional decision-making through real-time insights but expressed concerns about data security and the accuracy of AI-generated insights. The study outlines the design, implementation, and evaluation of the tool, highlighting its contributions to educational outcomes, practical integration within learning management systems, and future refinements needed to address privacy and accuracy concerns. This research underscores AI's role in shaping personalized, data-driven education., Comment: 26 pages, 11 figures, 8784 words

This paper presents a novel framework, Artificial Intelligence-Enabled Intelligent Assistant (AIIA), for personalized and adaptive learning in higher education. The AIIA system leverages advanced AI and Natural Language Processing (NLP) techniques to create an interactive and engaging learning platform. This platform is engineered to reduce cognitive load on learners by providing easy access to information, facilitating knowledge assessment, and delivering personalized learning support tailored to individual needs and learning styles. The AIIA's capabilities include understanding and responding to student inquiries, generating quizzes and flashcards, and offering personalized learning pathways. The research findings have the potential to significantly impact the design, implementation, and evaluation of AI-enabled Virtual Teaching Assistants (VTAs) in higher education, informing the development of innovative educational tools that can enhance student learning outcomes, engagement, and satisfaction. The paper presents the methodology, system architecture, intelligent services, and integration with Learning Management Systems (LMSs) while discussing the challenges, limitations, and future directions for the development of AI-enabled intelligent assistants in education., Comment: 29 pages, 10 figures, 9659 words

Counterfactual explanations for machine learning models are used to find minimal interventions to the feature values such that the model changes the prediction to a different output or a target output. A valid counterfactual explanation should have likely feature values. Here, we address the challenge of generating counterfactual explanations that lie in the same data distribution as that of the training data and more importantly, they belong to the target class distribution. This requirement has been addressed through the incorporation of auto-encoder reconstruction loss in the counterfactual search process. Connecting the output behavior of the classifier to the latent space of the auto-encoder has further improved the speed of the counterfactual search process and the interpretability of the resulting counterfactual explanations. Continuing this line of research, we show further improvement in the interpretability of counterfactual explanations when the auto-encoder is trained in a semi-supervised fashion with class tagged input data. We empirically evaluate our approach on several datasets and show considerable improvement in-terms of several metrics., Comment: 12 pages, 3 figures, 4 tables

Miscommunication and communication challenges between instructors and students represents one of the primary barriers to post-secondary learning. Students often avoid or miss opportunities to ask questions during office hours due to insecurities or scheduling conflicts. Moreover, students need to work at their own pace to have the freedom and time for the self-contemplation needed to build conceptual understanding and develop creative thinking skills. To eliminate barriers to student engagement, academic institutions need to redefine their fundamental approach to education by proposing flexible educational pathways that recognize continuous learning. To this end, we developed an AI-augmented intelligent educational assistance framework based on a power language model (i.e., GPT-3) that automatically generates course-specific intelligent assistants regardless of discipline or academic level. The virtual intelligent teaching assistant (TA) system will serve as a voice-enabled helper capable of answering course-specific questions concerning curriculum, logistics and course policies. It is envisioned to improve access to course-related information for the students and reduce logistical workload for the instructors and TAs. Its GPT-3-based knowledge discovery component as well as the generalized system architecture is presented accompanied by a methodical evaluation of the system accuracy and performance., Comment: 26 pages, 7 figures, 5 tables

Nonspecific molecular adsorption like airborne contamination occurs on most surfaces including those of 2D materials and alters their properties. While the surface contamination is studied using a plethora of techniques, the effect of contamination on a confined system such as nanochannels, nanopores leading to their clogging is still lacking. We report a systematic investigation of hydrocarbon adsorption in the angstrom slit channels of varied heights. Hexane is chosen to mimic the hydrocarbon contamination and the clogging of the angstrom-channels is evaluated via a Helium gas flow measurement. The level of the hexane adsorption, in other words, the degree of clogging depends on the size difference between the channels and hexane. A dynamic transition of the clogging and revival process is shown in sub-2 nm thin channels. Long-term storage and stability of our angstrom-channels is demonstrated here up to three years, alleviating the contamination and unclogging the channels using thermal treatment. This study highlights the importance of the nanochannels stability and demonstrates self-cleansing nature of sub-2 nm thin channels enabling a robust platform for molecular transport and separation studies. We provide a method to assess the cleanliness of the nanoporous membranes, which is vital for the practical applications of nanofluidics in various fields such as molecular sensing, separation and power generation.

GQA~\citep{hudson2019gqa} is a dataset for real-world visual reasoning and compositional question answering. We found that many answers predicted by the best vision-language models on the GQA dataset do not match the ground-truth answer but still are semantically meaningful and correct in the given context. In fact, this is the case with most existing visual question answering (VQA) datasets where they assume only one ground-truth answer for each question. We propose Alternative Answer Sets (AAS) of ground-truth answers to address this limitation, which is created automatically using off-the-shelf NLP tools. We introduce a semantic metric based on AAS and modify top VQA solvers to support multiple plausible answers for a question. We implement this approach on the GQA dataset and show the performance improvements. Code and data are available in this link \url{https://github.com/luomancs/alternative_answer_set.git}., Comment: accepted to EACL 2021

Future of sustainable fashion lies in adoption of AI for a better understanding of consumer shopping behaviour and using this understanding to further optimize product design, development and sourcing to finally reduce the probability of overproducing inventory. Explainability and interpretability are highly effective in increasing the adoption of AI based tools in creative domains like fashion. In a fashion house, stakeholders like buyers, merchandisers and financial planners have a more quantitative approach towards decision making with primary goals of high sales and reduced dead inventory. Whereas, designers have a more intuitive approach based on observing market trends, social media and runways shows. Our goal is to build an explainable new product forecasting tool with capabilities of interventional analysis such that all the stakeholders (with competing goals) can participate in collaborative decision making process of new product design, development and launch.