905 BYOD: How to Build a Simple Machine Learning Model from Virtual Reality xAPI Data
10:00 AM - 11:00 AM Thursday, June 27
VR and AR can generate a great deal of data, and that includes data in the xAPI format. One promise of this data is that it can be used to create predictive models that can classify learners; helping L&D departments to better serve them with appropriate content.
In this session, you will get your feet wet combining xAPI data from VR experiences and a machine learning model to gain new insights about learners. You’ll start with a guided tutorial for how to build a simple machine learning model using Google's Tensorflow. You’ll then use this example, with pre-supplied data, to understand how xAPI can be used to create predictive models. You’ll also investigate how best to generate xAPI so that it works well for model creation.
In this session, you will learn:
- How machine learning really works
- How machine learning can add value to learning in VR
- How to build a simple machine learning model in Tensorflow
- How to design and prep xAPI data for machine learning
- An introduction to different kinds of AI models (e.g., supervised learning, CNN's, RNN's, etc.)
- An intuitive understanding of how to use AI in the future
Designers, developers, managers
Technology discussed in this session:
xAPI, Tensorflow, AWS Sagemaker, IBM Watson, Unity, artificial intelligence, machine learning
Technology Required (BYOD only):
Hugh Seaton is GM of Adept Reality, a software company focused on using VR/AR in adult learning. Prior to Adept, Hugh founded AquinasVR, a VR/AR software company which he sold to the Glimpse Group, parent of Adept. Hugh’s focus, whether in immersive technologies, IoT or artificial intelligence, is on the intersection of learning science, creativity, and the cutting edge technologies that can bring learning to new levels of effectiveness.