Phoenix Perry on Interact ML
Phoenix Perry was our latest guest during the VR Design Research Labs, talking to us about environment conscious data, new ways of thinking and Machine Learning.
Our guest, before talking us through the Interact ML project she is working on, introduced us to the topic of some issues related to data processing in the modern world.
What we have fed in the system, stays there, and can have catastrophic consequences in terms of biassing and discrimintaion of people, groups, sexes, nations etc.
Training AI models uses up a lot of energy and in the light of a climate emergency, data has to be planet, environment friendly. Phoenix represents an approach of data being an open source, so available for free. At the same time, she is against globalisation of data, which means that there is no need for all people to have access to the same data – we don’t have to have everything universal for everyone.
Next, Phoenix talked about sensing and perceiving the world which is not actually human-centred. We just mistakenly think that everything should be focused on humans and their comfort.
The way of thinking and playing games relates to the way we perceive the world in general. Pragmatic action is an action that changes the world, Epistemic action is an action that changes the nature of our mental task.
Therefore, the idea of Interact Machine Learning, using the Unity node system. Interactive Machine Learning allows non-experts to create and test Machine Learning models. Record, train and run as three pillars of the project.
Interact ML is a visual scripting for Unity. It enables game creators to develop games without traditional scripting (writing code)! ML models are created by joining nodes in Unity together and visualising the data in real-time. It is a tool that helps people to add movement interactions to their Unity projects. First, the user needs to train the Interact ML the movements through performing them. Next, can instantly use those movements to trigger desired events.
Interact ML provides a range of ready models using various input devices, it is available to download for free from GitHub. There is also the Discord channel, which we are welcome to join, as well as a mailing list.
I am quite excited about the Interact ML and would like to give it a go in the near future. It would be great to implement it in the jigsaw puzzle VR experience I am working on. I loved the demo and the little teaser I watched on the project’s website. I have already joined their mailing list.
Phoenix creates physical games and embodied experiences. She also advocates for women in game development and established Code Liberation Foundation, teaching girls and women programming games. Social change through games and enriching industry with new voices are her other areas of interest. Phoenix currently works as an MSc Creative Computing Course Leader at UAL.