Innovators of Tomorrow | Justin Shenk

  • Overview
  • Transcript

This episode we feature Intel Software Innovator Justin Shenk whose projects focus on interactive applications such as visualizing neural network training, emotion detection games, and more.

Follow Justin Shenk and his projects on Dev Mesh

Checkout the Innovators of Tomorrow YouTube Playlist

Subscribe to the Intel Software YouTube Channel

Welcome to Innovators of Tomorrow. I'm your host, Wendy Boswell, here to bring you the sights, sounds, and inspirational work from developers in our innovator community from around the globe. Today, we are talking to Intel software innovator Justin Shenk, whose projects focus on interactive applications such as visualizing neural network training and emotion detection games. Let's get this started. 

Hey, Justin. Thanks for calling in from Stockholm, Sweden. You've been doing a lot of work with artificial intelligence. Meet ups, workshops, all that kind of good stuff all over Europe. Can you tell us what got you into this space and what your background is? 

Hi, Wendy. Thanks for having me on. My background is in Neuroscience research. I was looking into how memory emotion and speech disorders can be understood at the level of the brain. I've been working with some open source projects since then, like open [? TV ?] and it encouraged me to learn Python and Java at home. And I joined a study program a couple of years ago at the University of [? Osterbrook ?] in Germany, focusing on AI, specifically machine learning. And I happened to find that much less complex than the brain. 

All right. So that's an intriguing comment. Can you give us more information on what you mean, less complex than the brain? 

The brain is very complex and has many different parts that are interacting, and drawing a line between them the same way you can do with hardware or software is not possible. The machine learning technology on the other had is able to find some of the patterns that where once thought only possible to be done by the brain. So having the ability to detect patterns with technology like machine learning is quite attractive at the moment. 

That is really interesting. OK. So Justin, tell us a little bit more about the projects that you are currently working on. 

The biggest project I'm working on at the moment is master thesis, which is the Swedish company called [? Plotarian. ?] And their main project is a deep learning platform that's collaborative and has a graphic user interface. 

OK. Tell us a little bit more about this thesis project. 

I'm trying to make the internals of a neural network training and the process to make that intuitive and visible to people who are using it. So, for example, if you were wanting to classify an image as a cat or dog you could use a system, a neural network system, that would do a lot of heavy lifting in the background while still communicating to you about what you could do to improve the performance of your neural network. 

All right. So that sounds like a really cool project that could have some great real world applications. I also know that you put together a meet up recently around International Women's Day. Can you tell us a little bit more about that? 

Yes. I recently organized an International Women's Day celebration event in [? Rottenburg. ?] And the goal was to recognize outstanding work of women in AI research. 

OK. So tell us a little bit more about this event. How did it go? How did you organize it? 

So I invited local women at my University. and in the area who had demonstrated research in AI. And would be relevant in the wider community. So they came in and I gave a talk about the research. And I was able to get support from [INAUDIBLE] 

That is so cool, Justin, that you're supporting diversity inclusion events in your area around artificial intelligence. Is there anything else that you're working on that you'd like to tell us about? 

Sure. One of my favorite hobbies right now is developing a web app for emotion detection game using [? computers. ?] It was recently awarded the Open Source Software prize at my University. 

Cool. Can you tell us a little bit more about this emotion detection game? 

So the game takes a picture of players and compares the emotions that they show to whatever the prompt is. So if the prompt says show happiness, then the players will be ranked based on how they showed that. If it asks show surprise, then the players will be ranked based on their ability to show surprise. 

Oh, that's really cool. So how are people responding to it? 

So over the past couple of years, I've had the chance to demo the game at the Institute of Cognitive Science Christmas party. And it's been rewarding to see people interacting with it in ways that I wouldn't have expected. So, for example, making funny faces the whole night long. What I think maybe is particularly meaningful about it is that there is a possibility of using it in education for people with autism. So [? I ?] did have a relative who was diagnosed with Asperger's syndrome, and I would find it very meaningful to see it used [INAUDIBLE] who struggle to recognize [INAUDIBLE] emotions. 

Yeah. I think this could really have a lot of real world applications for people who struggle with facial recognition on the autism spectrum or people with developmental disabilities. I think that could be really super helpful. So Justin, how do you see your projects helping people in the future? 

Well, the majority of my projects are focusing on making AI algorithms interactive and visual. So I'd be quite happy to see them used for education, specifically to make them more accessible to a wide range of populate rather than [INAUDIBLE] specifically is a really useful field for this kind of work because the features are intuitive and people can get direct feedback about how their activities [INAUDIBLE] processed by an algorithm. 

Justin, that is really cool. I like how you're using AI and your projects to solve real world issues. I think that's going to be really impactful. Thanks for calling in today. We really appreciate it. 

Thanks a lot for having me on, Wendy. 

You can connect with Justin and follow along with his projects at the link provided. Also, we've included a link so that you can learn more about the Intel Software Innovator program. That wraps up this installment of the show. Be sure to like this video and subscribe to the Intel Software YouTube channel to keep learning about the innovators of tomorrow. On behalf of an amazing video crew, thanks for tuning in and we'll see you next time.