Andreas Refsgaard teaches workshops for creatives, artists, developers, coders, interaction designers as well as people with no prior experience with code or machine learning.

Andreas Refsgaard teaches workshops for creatives, artists, developers, coders, interaction designers as well as people with no prior experience with code or machine learning. Through fun, artistic examples and hands-on exercises Andreas encourages his students to play and experiment with the tools and techniques, thereby opening for new ways of producing ideas, developing creative projects, optimising workflows and critiquing the shortcomings of artificial intelligence and exposing its limitations. The workshops vary in length from a few hours to days or whole weeks and can be tailored to the needs of the specific team or class.

Machine Learning and AI for creatives

In his workshops for creatives, artists and designers Andreas demonstrate the creative potential of machine learning, code and algorithms. The workshop program often includes a basic introduction to the techniques and tools as well as various hands-on exercises where the participants get to experience the many ways in which the tools and techniques can contribute to the creative process. On longer workshops the students will also get time to work on their own projects, which will be presented at the last day. Andreas has previously taught workshops on creative coding and creative use of machine learning at companies like IDEO San Fransisco and educational institutions like ECAL, Umeå Institute of Design, Hyper Island, The Bartlett and SchoolOfMa.

Generating Images from Text

This workshop is ideal for individuals fascinated by the intersection of text and image, made possible by techniques such as Midjourney and Stable Diffusion. The course provides an overview of how these innovative AI tools allow for the generation of images from textual descriptions. Andreas demonstrates the potential applications of these technologies, showing their utility in fields as diverse as artistic expression, design prototyping, or commercial product development. The participants will engage with interactive examples and hands-on exercises that will help them understand how to employ these techniques in their own creative endeavors.

Tools taught include Midjourney, Stable Diffusion, ControlNet, DALL-E, RunwayML and other cutting-edge generative models.

Large Language Models (ChatGPT)

The workshops focusing on Language Learning Models are designed for both novices and seasoned artists, creatives and developers interested in the practical applications of AI-powered text generation, such as ChatGPT. Participants are guided through the basic applications of these models and their role in creative and utilitarian projects. Andreas takes an engaging approach, demonstrating with interactive examples how these models generate human-like text and how to customize them for unique uses. Participants then have the chance to try hands-on exercises using platforms like Hugging Face and OpenAI's GPT models.

Basic introduction to Machine Learning and AI

The basic workshops are aimed at people with no prior experience with code or machine learning, who are interested in getting acquainted with the technology and getting concepts such as AI and machine learning demystified. In the workshops Andreas gives a basic introduction to the field and demonstrates some basic and easy to use tools and techniques. The participants hereafter get to experiment with these through hands-on exercises on their own computer. The workshops typically conclude with a discussion about the new possibilities and ethical dilemmas, which arise with the increased influence of machine learning and AI in our everyday life. Andreas has taught workshop in basic machine learning and AI at libraries, public schools and highschools around Denmark.

Tools taught include Teachable Machine, ml5 and various demos from Google, NVIDIA, OpenAI, HuggingFace etc.

Machine Learning and AI for developers and coders

In his courses for creative coders, web developers and interaction designers, Andreas provides an introduction into how machine learning models can be productively integrated into existing workflows, integrated into websites and used for professional client projects. Through creative examples and hands-on exercises, he rolls out the many creative and technical possibilities machine learning brings to the field such as offering new and effective ways of prototyping and testing ideas. Andreas has previously taught workshops in machine learning and AI for creative technologists and developers at Facebook Reality Lab and at conferences like KIKK and Resonate.

Tools taught include ChatGPT, RunwayML, ml5js, Face-API, p5js, openFrameworks, tensorflow.js etc.

Recorded Workshops