If you spend time in wooded areas, you probably know how important it is to check for ticks afterward. Join us to learn about the Tick App, a citizen science project developed by researchers at UW–Madison with collaborators across the U.S. In this informative talk, we’ll show how the Tick App enables users to submit pictures of ticks they’ve found and how researchers identify what kind they are. We’ll discuss the evolution of this identification system and how we’ve started integrating AI into it.
Attendees will:
Discover how the Tick App enables users to submit pictures of ticks they’ve found and contribute to valuable research
Learn about the semi-automated, human-in-the-loop identification process used by researchers to classify tick species
Gain insights into the evolution of the Tick App’s overall system and the lessons learned during its development
This presentation will guide you through running a freely available large language model (LLM) locally on your own hardware. You’ll gain practical experience making your own data accessible to the model and explore various techniques to enhance its performance. The session is designed to provide a deeper understanding of LLM capabilities and limitations through hands-on demonstrations.
Attendees will:
Gain a clear understanding of Large Language Models and how they work
Learn how to set up and run the Llama2 model on your local Macbook
Discover how to leverage the LangChain library to interact with and extend the capabilities of LLMs
Understand the importance of system prompts and how to craft them to guide the model’s behavior effectively
Explore techniques for maintaining context and coherence in conversations with the model
Learn how to enhance the model’s responses by integrating relevant information from external sources
Discover how to represent and organize your data using embeddings to improve the model’s understanding and performance
This session will be a presentation where the use of AI in educational settings is demonstrated in order to spark discussion on the ethical use, limits, and guidelines for AI in educational settings. The session will involve using Free AI text an image generators to build a presentation on the ethical use of AI in education, followed by giving the presentation, and then a discussion.
This session provides a high-level overview of artificial intelligence (AI), covering its history, concepts and various applications. Attendees will gain a deeper understanding of the different types of AI, using illustrated examples to demonstrate how these technologies work and explore the potential solutions and challenges associated with the effective implementation of AI. By the end of the session, you will have a clearer understanding of AI, its various types and uses and the successes and challenges you may encounter when implementing AI solutions.
Attendees will:
Learn the definitions and types of AI
Understand the history, concepts, and paradigms of machine learning (ML)
Explore deep learning, artificial neural networks, and generative AI
Discover the potential solutions and applications of AI technologies
Identify and discuss the challenges and consequences of using AI effectively
Hear how DoIT’s Research Cyberinfrastructure unit is harnessing the power of AI to open new avenues in data analysis, including for meteorological fact-finding. Discover how cutting-edge large language models (LLMs) like OpenAI’s ChatGPT 3.5 and 4.0 hold are poised to revolutionize the field of meteorology.
Attendees will:
Explore a novel methodology that leverages LLMs to generate tailored analytical code for specific meteorological data sets
Witness a compelling case study comparing findings from internet sources and LLM-generated analysis
Learn how ChatGPT 4.0 can create reusable programs for data analysis, drastically reducing time and effort in coding and debugging
Gain insights into the challenges and opportunities of utilizing LLMs in meteorological fact-finding
Discover how the School of Medicine and Public Health (SMPH) implemented a secure data enclave using DevOps, Azure cloud computing and Microsoft’s suite of security tools. SMPH’s platform enables researchers, data engineers and data scientists to utilize resources like virtual machines, databases, OpenAI, data lakes and Power BI, all while protecting patient privacy. Gain insights into creating and scaling a secure computing environment in the cloud and learn about the platform’s capabilities.
Attendees will:
Learn about the approach SMPH took in creating their secure enclave
Hear about the various resources available within the platform
Discover how the platform maintains patient privacy while providing access to powerful tools for research and analysis
Gain practical tips about how to create and scale your own secure computing enclave in your IT environment
