In this episode, I am talking to Dale Lane, the creator of Machine Learning for Kids – an extension to Scratch that allows kids to train their own ML models and use them from their Scratch programs.
As Dale writes on his website, he’s a father of two exhausting kids, a software developer for IBM in the UK, a mobile and gadget obsessive, a charity trustee and an all-round geek. Dale is with IBM for more than 16 years now. During his career, he came across many of IBM’s Artificial Intelligence and Machine Learning offerings, such as IBM Watson. While his work at IBM most definitely prepared him to build something like Machine Learning for Kids, this educational offering was built entirely in his free time.
So what exactly is Machine Learning for Kids? It allows kids to train their own machine learning models to recognise text, numbers, images, or sounds. Just years ago this was only possible for developers with specialized knowledge – making this learning available to kids is absolutely phenomenal. The trained models can later be used in Scratch, which then combines Machine Learning with visual, block-based coding. After integrating with Scratch, Kids can simply drag visual code blocks into the editor to make use of the machine learning models.
Machine Learning for Kids started as a coding tool that Dale created for local schools near his home in the UK and now has turned into one of the most used IBM Activity Kits ever. In 2018, Dale was awarded the IBM Volunteer Excellence Award—the highest form of volunteer recognition given by the company.