Siddhartha Srinivasa

Women@SCS conducted an interview with Siddhartha Srinivasa, Finmeccanica Associate Professor in Computer Science.

Photo credits go to Jim Judkis

To start off, can you tell us a bit about your background and your journey before coming to Carnegie Mellon?
I’m from India. I grew up in Chennai by the beach -- that’s where I spent the first twenty years of my life. I grew up in a university campus in the Indian Institute of Technology at Madras. My father was a professor, so I was told that I should be a professor since I was five years-old. I’m glad I didn’t disappoint people too much [laughs], but yeah, I grew up in a very academic background. I think this idea of asking questions was something that was very fundamental to everything I did, and I remember fondly dinnertime conversations with my dad where he would just ask me why, why was something true. I think that was a good start to becoming an academic. I did my undergrad at the same school, the Indian Institute at Madras, in mechanical engineering. I studied fluid mechanics, and then I came here to do robotics because I thought robots were cool.

What attracted you to Computer Science in the first place?
Well, I think it was very serendipitous, the whole thing. I had a background in mathematics growing up. My dad bought a computer when I was seven years-old, and I started programming with it (and playing games also). I started writing my own video games and other things, basic things, it was really straightforward. But I learned a lot of programming that way, by writing my own code, so by the time I was 13 I was fairly proficient at programming. I was sort of ambivalent to it, but I think robots particularly captured my imagination because it’s the manifestation of algorithms and computing in the real world. That’s what I love about robots -- they're a way to extract human instinct. I watch people pick up coffee mugs and I think, why do they do what they do. You write an algorithm about it, then you put it on your robot and sure enough, it does the same kind of interesting, funny, funky motion that you did. It’s almost like I captured some piece of human behavior, and it feels good. We’re slowly uncovering this big, giant wonderful puzzle of how we, as humans, naturally and seamlessly interact with the world. It’s also impossibly hard, so it’s a good challenge. Building robots that can have human capabilities is something that we’re all striving for, and hopefully we’ll get there eventually.

What’s your favorite part of your job here?
I love everything. My favorite part of my job here is working with students. I think it’s a very odd feeling, because I get older, but the students I interact with stay the same age or get younger. I came here 17 years ago and, some of my first PhD students were about my age. I was 19 when I came here, and now they’re young enough to be my children. It’s one of those things, you know, that keeps you young. I think that it’s also really amazing to watch them mature and grow from these people who don’t know what a robot is into these brilliant young minds. I have 2 young children, 5 and 2, and I have the same feeling with them that I get every time I work with my students, just like, “Oh my god, I can’t believe you figured this out. That’s so adorable!” I think being able to work with students and, in whatever little way that I can, helping them learn about being scientists and helping them learn the right sort of way to solve problems is very satisfying.

What would you say is the most challenging part of your job?
Probably the same thing! [laughs] You know, I think it’s a lot like having children. The highs are really high, the lows are really low. I think if there’s anything challenging, it is that I have really high expectations of my students. I really think that they are, and I want them to be the best in the world at what they do. Just like all of our parents have high expectations of us, being an academic advisor is very similar to that. And so I think one of the things that I care a lot about is, I think we should do whatever it takes to solve the problem. Solving the problem is like 70%, “Oh my god, it’s so fun” - But the other 30% is just putting your head down and just working as hard as you can. There’s nothing that beats hard work, so sometimes it frustrates me that that 30% isn’t done with the same excitement as the other 70%. But I think some of that perspective comes with age, like 17 years ago I was a very different person than I am now. I probably would’ve hated me as a student. But I think that’s the challenge, is just making sure that the expectations are met and that the standards are kept. Essentially in our field the only thing we have is our reputation. People look at your papers and think, ok, it’s going to be good because it has your name on it, so maintaining that high quality is something that’s hard and challenging.

What achievement are you most proud of?
When I started as a professor, I thought: I want to build a robot from scratch that can autonomously perform the kinds of tasks that I perform at home in my kitchen, like clear the table, make a meal, separate the cookie from the cream of an oreo, and I was very determined to make that happen. No such system existed then. There was no existence proof; nobody knew it could be done. And I think by a lot of hard work and a lot of luck, we built the system.

It took a long time, and we’re still building and rebuilding it. Herb, my robot, can actually perform a lot of very complicated manipulation tasks in a reasonably challenging environment. So really being the first to show that you can do these tasks in the real world has been very gratifying. I think that it has also encouraged others to try and solve this problem, because you need one example and I’m actually in many ways very happy when someone beats me at a particular task, because at least I got them excited about trying to solve this problem. So I think that from very early on I was very committed to not only writing papers and proving theorems and writing algorithms but putting it out there in the real world so that we can build an end to end system, and I’m really happy that we were able to do that. It took many years and many wrong turns and I think we’re still refining that system but we definitely were able to do something that no one else in the world was able to do. It was great.

Could you tell us a little bit more about the the current research projects that you’re working on?
Sure, one of the topics that I’m really interested in is assistive care. I want to build robots that can actually help people. I think we spend a lot of money and a lot of effort in producing technology that can make able-bodied people more able. We have phones, watches, Fitbits, all of these things, but there are over 6 million people in the United States alone who have disabilities of their upper extremities, their arms as well, and to them, even if they drop their TV remote they need to call a caregiver to help them. So I’m really excited about a new project that we have where we’re trying to help people with high spinal cord injuries, paraplegics, quadriplegics, people who have very limited mobility neck down, to use robot arms that are mounted on their powered wheelchairs to perform complex manipulation tasks like pull out a meal from a fridge, microwave it, and eat from it -- and these are things that they just can’t do -- sort of enable that and be able to do that is both a very very challenging robotics problem, like how manipulating food is hard. A deformable object, you cut things and you twirl things and you pull things but it’s also a really wonderful human robot interaction problem. Like how do you like this arm that’s right next to you, like how do you want it to behave? Do you want it to behave like an extension of your own arm? You want it to like turn around and talk to you? I don’t know, right? We don’t know the answer to those, so it’s this, it’s a shared trauma. And if you think about the ways a caregiver takes care of somebody. Right? And it’s very very complicated and they’re not only assisting you physically, but they’re also predicting your intentions, they’re predicting your mental state, they’re predicting your emotional state, they’re caring for you. Care, you know is a lot different than they’re feeding you. So I think I want to give caregiving robots. Again, like before, this is something that has never been done before. I’d love to have something that can help my grandmother live an independent life for longer. So, it’s my next big challenge. I’m excited to take it. Also working very closely with the Rehabilitation Institute of Chicago, world’s best at rehabilitation as well as with a company called Kinova based in Montreal. We’re running several user studies with real patients with high spinal cord injuries. Interestingly, a lot of these people are kids. People your age who, bungee jumping or skydiving, fell and broke their spine and can’t move anything below their neck for the rest of their lives. And I think that what I find most amazing about them is that they’re filled with excitement. They really want to use this technology, they really want to change their own lives. So I think anything I can help them do will be great.

What’s your perspective on women in computer science?
I think it’s really important to think about gender bias and gender equality in engineering and in computer science, so I actually have been working really closely with HereMe which is a group that is part of the School of Computer Science. If you’ve seen these kiosks that are out there, they’re what we’ve been putting out mostly to understand the role of gender bias in STEM. A lot of this came about because Jess Kaminsky from HereMe and I were interviewing some kids from my sons preschool. He’s five and one of the kids pointed to her and said - you don’t build robots, he builds robots. And I was just shocked. I was like - why do you think so? What’s up with that? And of course I told him very strictly that that’s not true. But I think that it got me thinking and got us thinking about how can we explore this idea about bias in science. We went to various high schools in Pittsburgh and talked to them about science and STEM and whether they’ve experienced bias. And I think it was very rewarding - it was mostly to serve two purposes, one is to educate our own communities about what it means to be a woman in science and I have a 2 year old daughter and I want her to grow up to be a roboticist, obviously. (laughs) And I also want to make sure that I don’t hear my son say the things that I heard that kid say. But I think that it’s very complicated - we live in a society where there are implicit and explicit biases that people have about what roles that people play and it’s up to us also as a School of Computer science to think more than just our school but to think about the fact that we’re part of a community and it’s important to engage in that community. So I’m very excited that my lab is one of the few labs I think that they're about as many women as there are men, in terms of undergrads, in terms of PhD students. It’s not something that I consciously wanted, it’s just something that was built over time. I think having women role models means more women are excited about being part of the lab. We have several role models - one of my students is now a professor at Berkeley. One of my undergraduate students, Rachel [Holladay], is now a mentor for FIRST and is incredibly involved. I think that the more role models we have the more future generations will think that - oh it’s obvious that everybody can do computer science, right? I think that’s really important. And I think continuing on anything that I can do to make sure that we produce students who are not only technically competent but also socially and morally responsible and have a good head. I think it’s our duty to do that.

Could you elaborate on what you think the future holds for robotics?
I think the future is awesome for robotics. I’m really excited to be part of it. I think that we’re already seeing a proliferation of technology in our lives. Ten years ago, if I told you that you’d have a computer that was a phone and had a camera that was like the biggest camera you’d ever had, you’d never have believed me -- and that you’d be staring at it for most of the day, playing Pokemon Go. So it’s one of those things that I think technology has often evolved faster than we can ever imagine, but I think one of the challenges with technology right now is the final frontier is really embodied technology -- technology that’s not just sitting in your phone passively, but that’s walking around us doing things like robots that can actually do useful things and I think within our lifetimes we’ll have robots that are helping us with our everyday lives and that’s a huge inflection point. That’s when a lot of interesting things can happen. Particularly, I’m excited about care. I think that not only will we help people with spinal cord injuries and older adults -- there are so many people in the world now who need assisted care and the cost of caregiving is going up tremendously. I think technology can really help us and robotics can really help us enable people to live in their lives much longer. So I think the future is great, and I’m excited to be a part of it.

Are there currently any opportunities for undergraduate students to get involved in your research?
Absolutely! I have a lot of undergrads already and I would love to have more undergraduate students work with me. We have a lot of projects. We demand a lot of different skills from our students: there are mechanical engineers, electrical engineers, computer scientists, people interested in human-robot collaboration. We have a big lab with lots of students and for me, working with undergrads has been really rewarding. A lot of them have continued on to their phDs with me. A lot of them have continued on to do their phDs at other fantastic places. But, you know, robotics is one of those domains where the more hands-on experience you have, the more you work with a robot, the better you are, and more you learn. So we have two really nice robots platforms -- Herb and Ada -- and we’re more than excited to have people work with us. If any of you are ever interested in anything related to robotics, talk to me. We’re also really committed to outreach. We do a lot of outreach in our lab and we’re excited to have even more opportunities to do outreach, whether it’s working with kids from underperforming schools, or working with kindergarten kids, or working with schools from all over the world. I’m really excited about that and looking for more ways we can spread robotics material out in the real world. And we release open source code -- all the code that we write is completely open source so we can build an ecosystem for robotics.

What do you like to do in your free time?
I don’t have very much of it [laughs]. I run a lot. I’m training to qualify for the Boston marathon. I’ve run marathons and ultra-marathons, but five years ago, I started running quite a bit. I used to captain the CMU squash team -- I played a lot of squash. Before that, in India, I used to play a lot of tennis. And then I’m a full time dad. I have a five-year-old son and a two-year-old daughter and I like spending time with them. I also like to cook. I cook almost every day. That’s what I do, and I have an iPad -- I sketch on it sometimes, it’s fun. I try to make sure that when I’m not sleeping, I try to do something. I’ve been lucky to have very high energy and high metabolism.

How do you find time to balance all the work that you do and all the activities that you mentioned?
That’s a great question! I think that it’s important to say no to things that you won’t be able to do. I have really, really strict priorities on what I think is important. And I try to stick to them very, very hard. And I think whatever I take on I try to do 100% and so I make sure that I can do that. I also have a very organized calendar, to-do list, and email management. I’ve also learned that delegating things works a lot. I say no to more things than I say yes to, and that helps me maintain sanity. I also just block off time to do things. So I run a lot and it’s just blocked off in my calendar so I can’t not do that. So it’s nice. I think having kids really changed my perspective on things for the better. So, at 6pm I go home and I’m just not available to anyone other than my kids. It’s nice.