Gordon Erlebacher
Department of Scientific Computing,
Florida State University (FSU)
"AI Tools for Education and Research"
Sep 06, 2023
Abstract:
In the upcoming colloquium, attendees will gain a holistic understanding of the evolving landscape of AI in research and education, grounded in both historical context and future potential. We will provide a comprehensive overview of the development and significance of large-language models, highlighting the foundational aspects of the transformer architecture. This examination will elucidate the various applications of the transformer and distinctly showcase the decoder-only variant, which is a next-word predictor. The discussion will progress to practical tools to support the composition of papers written in Latex using Overleaf, a popular platform. We will then offer a comparative analysis of models such as ChatGPT, GPT4, Claude, and Perplexity.ai, detailing the optimal usage scenarios for each. A significant segment will be dedicated to AI's role in education. We will assess its utility for students and faculty in the classroom while addressing concerns about academic integrity and potential misuse. Further, we will delve into the practical applications of AI in coding, with a specific focus on Copilot's integration in Visual Studio Code and its implications for development efficiency. If time permits, we will list some creative applications of language models using tools such as Langchain. We will conclude with an outlook on the future implications of AI in the educational sector and its potential impact on student learning and interaction.
