Machine Learning for Interactive Art

Become familiar with the tools and the theory at the intersection of machine learning, digital art and interaction design.

Workshop Dates: August 10–14, 2020 (10 am–4 pm CEST, plus additional time for independent group work)

Enrol here!

Faculty: Andreas Refsgaard & Jen Sykes

What is Machine Learning?

A hands-on introduction to machine learning with a focus on creating your own artistic and interactive applications.  

When programming the behaviour of an interactive prototype, digital artists and interaction designers often rely on logical programmed relationships between an input and an output.

Machine Learning suggests a different kind of logic. Instead of relying on explicit sets of rules to determine a system’s behaviour, machine learning models learn by example.

Through the advancement of accessible toolkits, artists and designers can look for patterns within examples, creating their own rulesets. This process is somewhat like our own brain, processing behaviours that can be quickly explored. This provides a lot of new, exciting opportunities for designers, musicians and artists especially when working with complex information often too complex to account for via coding.

This workshop will explore the basics of machine learning in a way that is tailored to creative practitioners with no prior experience in machine learning. The workshops especially focus on how participants can train their own models to act upon custom inputs and assign output behaviours to them.

What will you learn?

The participants will be provided with a set of easy out-of-the-box tools for training their own machine learning models or using pre-trained models through the javaScript library ML5JS or the machine learning software RunwayML. Because machine learning allows the participants to handle really complex inputs and easily map it to desired outputs, this workshop will:

  • Provide participants with lesser coding skills an alternative approach for making interactive tools and demos.
  • Be a hands-on approach to getting familiar with a complex, but increasingly important field.
  • Teach participants how to use machine learning methods to create real-time interactions for use in video, installation, live music performance or physical computing.
  • Allow participants to prototype faster across a range of mediums through examples and physical iteration rather than explicitly programming them. This process is more flexible and iterative compared to the more linear and time consuming programming approach.
  • Facilitate the creation of custom interfaces for this rapid prototyping, enabling a much more agile collaborative process.

What will you not learn?

This workshop places a focus on applying machine learning to creative disciplines. Whilst varying practical methods of working with inputs and outputs are explored, this workshop does not focus predominantly on analytical or statistical data within machine learning fields.  

How will you learn it?

Through:

  • Lectures
  • Small assignments + explorations
  • Group project with supervision

Is this for you?

Yes, if you are interested in exploring the world of Machine Learning and how to design for it. Anyone can join – prior coding experience (especially in creative coding frameworks like Processing, p5js, Arduino or OpenFrameworks) would be helpful but is not necessary to get the most out of the workshop.

What do you need to bring to the workshop?

  • A computer with a functional webcam and microphone
  • OPTIONAL: Any sensors, external webcams and microphones, etc. you might want to use as an input

Enrol now: There is a maximum number of 20 places available for each workshop, first come first served. Enrolment will be closed when the workshops are full.