 Well, good evening folks. My name is Ryan Hargraves and I am the Director of Admissions here at the University of Vermont. I wanna thank you for joining us here for this segment of UVM Live at Five. This Live at Five segment is an opportunity for you to get to know more about what the University of Vermont has to offer through the perspectives of some of our most distinguished community members. Today, we are very fortunate to be joined by a great faculty member, Professor Josh Bongard. And Josh has had a real impact on academic life and life otherwise at the University of Vermont. But I wanna thank him for taking time out of his schedule to join us. So why don't we, without further ado, give you an opportunity to introduce yourself, Professor Bongard. Sure, thank you very much, Ryan. And hello to all of you out there in cyberspace. As Ryan mentioned, I'm Professor Bongard or Josh. And I'm a professor here in the Department of Computer Science. I've been at UVM for about 10 years. And I came to UVM by route of a number of other universities. I grew up in Canada and I did my undergraduate studies at a university in Canada. And then carried on my studies into robotics and AI, which we'll talk about shortly, at a number of other universities in England, Switzerland, Japan, New York State, eventually got a job here at UVM. And in my travels, I was at five different universities. Some were primarily focused on undergraduate education. Some of those universities were very research intensive where the professors spent a lot of time in their labs and not so much time in the classroom. So when I was looking for a job, I was really looking for a place where I could teach and do my research. And I'm happy to talk about both of those. All right. Well, thank you for your introduction. Yes, so tell us about your work in the artificial intelligence space. What types of questions are you attempting to answer through your research? Sure, yeah. So again, my research group here at UVM, we work on robotics and AI. So generally trying to make smart machines. For a long time, this was something sort of that just scientists worked on in their lab, but in just the last few years, there's been some exciting advances in robotics and AI and everyone's involved in the conversation now. So it's an exciting time to be learning about robotics and AI and doing research in that area. The basic approach in my group is we use ideas from evolution. So if you think about our human ancestors, tens of thousands of years ago, we gradually domesticated wolves into dogs and we gradually domesticated the taosente plant into what today is corn and we did it through a very simple process. Plants and animals that provided some use to humans. We allowed them to reproduce and evolve and those that didn't help, we didn't allow them to do so. So in my group, we study this idea of evolutionary robotics where we use this ancient technique that our human ancestors discovered, but we apply it to a very modern technology. So here at UVM on our computers, we have virtual worlds inside our computers and inside those virtual worlds, virtual robots try and do useful tasks and those that do a better job at those tasks survive and reproduce and evolve and those that don't are deleted by the computer. So we evolve robots in our computers and in some cases, we then hook up our computers to 3D printers and print physical versions of these evolved robots and that's our approach to try and create intelligent machines. Well, that's some really cool stuff that you're describing there. And so, hypothetically, if I am a student looking at University of Vermont and I'm interested in STEM or your area, what advice, what guidance might you give me as I approach opportunities here? Sure, so whether you come to UVM or whatever university you end up, my primary advice is go knock on the doors of your professors if you're in their classes. After class, ask them what they do in the same way that Ryan was just asking me and you'll find that most professors are more than happy to tell you what they're excited about and what they work on and that's the best way to get started. Start conversations with your professors. In a lot of your classes, you'll find professors will come to other professors' classes and sort of advertise their research in the same way that I just did. So find out what your profs work on. Check out the department websites. Talk to them after class. Talk about what their other colleagues are doing in the department. Get the conversation going and that's usually a first step to opening the door and getting admittance to a research group on campus. Awesome. So could you tell us, you know, you said you've been here for a decade or so. So could you tell us about a specific student who's been able to really maximize opportunities, someone who's done something particularly interesting, a student who's really exemplified, you know, the opportunities available in STEM here at University of Vermont? Sure, I'd be happy to. I have one undergraduate student who's currently working with me, Jack Filag. Jack's currently a senior, but two years ago, when he was in his second year here at UVM, he found out about my work and knocked on the door of my office and sat down and asked me, again, questions like Ryan is asking and got really excited about what I work on. Jack's a mathematics student, but he took my course and built up his skills in robotics and AI, and then he joined a particular project going on in my lab called Twitch Plays Robotics. And in this project, we stream our evolving robots to the web where people can try and teach these robots various tasks through language. So it's a project that connects together robotics, language, crowd sourcing, the web, big data, mathematics. It's a very exciting project. Jack got really excited about it and he helped us actually build up the infrastructure to build the system to be able to allow people to teach robots through language. And Jack is a senior now and he's thinking about coming back to UVM as a master student to continue on his studies in robotics and AI. Wow. Well, that's pretty significant research, particularly for someone who is an undergraduate. So that's really cool to hear about that. But I have a serious question. You talked about some of the power that can be harnessed through robots. So my question is, should we fear robots and will robots at any point take over the world someday? That's a very good question and that, as you can imagine, comes up quite a bit for those of us that actually work in robotics and AI. As I mentioned, robotics and AI are becoming more and more popular. There's more and more drones out there, more and more discussion about autonomous vehicles. Should we allow them? How should we allow them on our roads? So as the technology improves, we're starting to see robots and AI that work literally alongside people here in the real world. And that raises very important questions about how do we make sure that these machines are both useful and safe at the same time? My answer to that is we need to make sure that these machines are able to understand people. And unfortunately at the moment, our machines are not very good at that. Some of you might have experiences where your phone rings in your pocket at the worst possible time. Your phone doesn't know that this is not the right time to send out an audio alert. Our machines in the world here at the moment don't have a good sense of people and what we like and what we dislike. So a big technical task, big job that we still have to do is how do we teach robots to be more social? How do we get them to understand that they're not alone in the world, that there are other people out there and they have to be careful about how they choose what to do and what to not do. That's the challenge. No one has a good answer to that at the moment and that's where research comes in. We need to try and study how do people become social? How do people understand how to get along in the world with other people? And how do we take those ideas and bring them back into robotics? So that's actually a connection between what's known as STEM, science, technology, engineering and mathematics, the technical sciences if you like and it connects with the world of people. So now in AI and robotics, if we wanna make safe machines, there isn't going to be a purely technical solution. We're gonna have to come up with a socio-technical solution which is just a fancy way of saying, we're gonna have to think not just about machines but about people and machines and how those two groups should best get along. So if I hear you correctly, basically the challenge is to give or facilitate the emotional intelligence or something equivalent to that in these machines. Is that kind of the quandary? That's a great summary. So we're getting very good at making machines that have higher IQ or book smarts if you like. We're not so good at making machines that have emotional EQ, emotional intelligence. Exactly, that's a great summary. That's a big challenge and that's something we work on here in my group. And if you decide to join us here at UVM, you may find yourself part of that adventure. Awesome. Well, I have met you in person and I can confirm that you are not a robot and as such to give the audience just an opportunity to get to know you better. What are some things that you, and I've asked this question of other participants on this segment, but what's something that you like to do with your free time? Ah, yes, that's a great question. Yes, I am definitely human. You can ask my students about that and they'll confirm that. For those of you that have visited Vermont, you know that it's a beautiful place. We have the beautiful green mountains nearby. One of my favorite things to do is to go snowboarding in the local mountains around here. I like to go on a day that's relatively quiet. Vermont is a small place. You can often find a ski hill that where there is just you in the mountain. And I love to sit at the top of the mountain and sort of think my thoughts, think about research, think about life. Vermont is a great place to think hard and work hard, but also to play hard and enjoy life. So that's one of the things that I like to do on my time off. Awesome, awesome. Well, professor, we wanna thank you for taking time out of your very busy schedule to join us here on this segment of UVM Live at Five. So we hope that you all enjoyed watching our interaction and stay tuned for the next episode of UVM Live at Five. Until that point, have a great evening and take care, everyone.