How can you find and pursue undergraduate research opportunities? Professor Alex Mariakakis shares his insights. And we hear from three students, Anatoly, Sarah, and Arkaprava, about their personal experience.
[1:29] Professor Alex Mariakakis shares examples of undergraduate research projects he has supervised.
[2:46] Check out the Department of Computer Science's Undergraduate Summer Research Program.
[3:08] Anatoly shares his experience with the Summer Undergraduate Research Program in Astronomy and Astrophysics.
[7:38] Sarah shares her experience doing undergraduate research in Natural Language Processing (NLP) with Professor Suzanne Stevenson.
[12:58] Arkaprava shares his experience pursuing undergraduate research in Quantum Computing.
[15:43] Check out the Centre for Quantum Information and Quantum Control.
[16:44] Check out the Research Opportunities Program.
[19:32] Professor Alex Mariakakis shares some advice on how to communicate with faculty about their research and pursue opportunities to work with them.
Diane: As you navigate your classes and get to know faculty, you might start feeling the itch to learn more; to extend past the classroom and get involved in the research that professors are working on. Computer science research is tackling problems in so many fields. And there are many ways to get your foot in the door. So, let’s hear from students and faculty about their experience doing research together.
Mario: Hey Diane, what do you feel is something that really makes UofT a fantastic place to study?
Diane: That would have to be research. Within the department, we have such a diverse set of researchers working on all sorts of topics. And outside of the department, computer science skills lend themselves so well to other research areas.
Mario: Hm, I guess, I can see what you mean. But how does all this research help our undergraduate students? They’re just here to take courses, right?
Diane: They can do much more than that! Undergraduate students can pursue research, too.
Alex: I have undergrads doing all sorts of different things, I have some students who are just helping us with data collection, so administering user studies, making sure that survey responses are coming in, I have some people who are building mobile apps, so either doing Android or iOS development, I have people doing machine learning, so they're processing sensor data that's in spreadsheets, for example, and they're, you know, coding up Python machine learning models, really the whole spectrum.
Mario: That was Dr. Alex Mariakakis, a professor at the University of Toronto who runs the Computational Health and Interaction lab (CHAI).
Diane: Alex gave us some very concrete examples of what undergraduate students work on in his research lab.
Alex: I have like one student right now who's looking at blood pressure estimation. So he needs to take some time to understand, you know, what is blood pressure? How's it normally measured?
I have another student who's looking at menstrual health. So, she has to understand, you know, what are the different phases of menstrual cycle? And what are the different physiological things that people would care about when we study that?
Mario: There are many other kinds of problems that students work on, and techniques that they use, in their research. In this episode, we hear from three students who share their experience in both finding a research opportunity and then doing research in that area.
Diane: There are lots of programs at the university that support undergraduate research. For example, there is the department of computer science’s undergraduate summer research program, UGSRP. But you can find summer research programs across the university.
Mario: That’s how Anatoly, a third-year student in 2022-23, began his research journey in the Department of Astronomy and Astrophysics right out of first year.
Diane: Anatoly, what was involved in applying to the Summer Undergraduate Research Program in astronomy and astrophysics?
Anatoly: So aside from the general information that you have to submit, I needed to submit a statement of interest, which included questions like, why I'm interested in the program, what kind of projects I'm looking forward to, and a little bit of history about myself as well. I also needed to supply my transcript, which I think only included the courses from Fall, first year. And I also needed one reference letter for the program.
Mario: Wait, you needed a reference letter? How did you manage that after completing only one year at the university?
Anatoly: So, in my first year, my first semester, first year, I took a course MAT 240, which is algebra one. And I was really, really active in the lectures I participated a lot in the discussions that we had during lecture time, I answered questions, I gave like suggestions. And the professor noticed me, so the professor didn't just see me as a name and a grade as he might if I don't really interact at all in lecture and you don't come to office hours. But if you actually make your presence known to the professor, then they actually have something else to go off of aside from just a grade.
Diane: Now, you may be wondering why a computer science student might be applying to undergraduate research programs in something like astronomy.
Anatoly: A lot of the projects that were offered by the Astro SURP, were really just software development. There was one on web visualization. And this one I did, which was on graph databases.
So, most of the work I was doing while there was a lot of theoretical astrophysics background that did go into it, I was really only working with the software development side of things.
Diane: So, your background in CS was really helpful to this project?
Anatoly: Oh, yeah, absolutely. One of the major factors that got me accepted for this project was my experience with software development.
Mario: You may not be fully satisfied with that answer. Surely, some domain knowledge in astronomy is required. But remember that a lot of that domain knowledge can be learned through your research supervisor.
Anatoly: I worked a lot with Professor Adam Hanks, who is Assistant, an assistant professor at the Department of Astronomy and Astrophysics. And I worked very closely with him when designing the API and all of the code. We had weekly meetings where we would just talk about what is the best way to implement this or implement that. The API that I wrote was loosely based on a previous API that Professor Hanks wrote in the past, so he had a lot of ideas and a lot of thoughts that absolutely helped me.
Diane: This varies. Sometimes, the supervisor will be a graduate student. Either way, there may also be a group that gets together for activities like discussing research papers, or presenting their work to each other. In the Department of Computer Science program, UGSRP, there are weekly meetings – and also social events.
Mario: One of the best things about studying computer science is that it intersects with so many other interests you might have outside of computers.
Sarah: …I've always been interested in creative writing. And I thought that may be some, there might be some kind of connection between computer science and creative writing. I was also taking linguistics courses in my first year for breadth. And I thought they were interesting. So that was sort of another connection there.
Diane: That’s Sarah, a second-year student in 2022-2023, who didn’t get into research by applying to an existing program. In her case, asking questions and talking to people opened doors.
Sarah: ...in first year, towards the start of the semester, I emailed the undergrad, computer science undergraduate department, for assistance on structuring my degree. I was looking into the NLP focus that U of T offers. And I was a little bit confused about what kinds of linguistics courses I would have to take, for example. So they actually forwarded my email to Professor Stevenson who I spoke to about this academic issue, but also about her research. And that got me interested in doing it.
…we met over a Zoom call. And she also introduced me to some of her past and current graduate students at the time. So I got to speak with them about their undergraduate course structures, and also about the research.
Mario: What started out as asking for help with course selection led to a conversation about research that, a few months later, really paid off.
Sarah: When I was actually looking for research opportunities in December or January, I reached out to professor's students, and again, just asking whether she had any positions open. And from there, she said, Yes, they were considering taking some, like a volunteer or something for this new project.
Diane: As part of the research project, Sarah had to do some reading about the research area itself.
Sarah: So initially, I had to read a bunch of papers, sort of to get context on what's going on in this field in general, I guess, or what, what existing research is relating to the work that we're doing right now.
Sarah: … for the first few weeks, it might have been like one or two papers a week, after a certain point, I just started doing more of like the computational stuff to start dealing with the data.
Mario: In Sarah’s case, she worked closely with a graduate student in Professor Stevenson’s group. We asked Sarah to explain a little more about what that work looked like.
Sarah: First, what we did was we created some visualizations with bar graphs sort of showing the difference in certain contexts. So you'd have subreddit and then you'd have the contextual contexts or contextual situations, like specific singulars, specific plural, stuff like that. And then the bar graphs would have the rate of gendered versus gender neutral. So then we'd see that there'd be some differences between the sub Reddits. And then after that, we did the regressions to see whether or not the, what we found was actually, like, statistically significant or not.
Diane: Some of what Sarah described to us requires a background in linguistics. But let’s focus on some key points here: data was gathered, in this case from a subreddit, and then some computation was performed on that data in the domain of linguistics.
Mario: If this sounds somewhat familiar, it may be because we do this quite a bit in computer science assignments, especially in first year. Many, or even all, of the assignments in courses like CSC110 and CSC108 are applying concepts from computer science to a cool problem.
Diane: Of course, in a course assignment, the problem domain is chosen for you and the questions are phrased in such a way that your answers demonstrate important learning objectives. But in research, there is no assignment handout. So, the visualizations and analysis done by Sarah and her research group not only provide answers but may even result in more questions.
Mario: Naturally, as educators, Diane and I asked Sarah which courses helped prepare her for this research project.
Sarah: …definitely 110 and 111 helped because I learned Python. And basically, everything that I'm doing for this project is written in Python. I would say the first few linguistics courses that I took, specifically LIN 102, also helps, because in that course, I sort of got an introduction into, like, parsing and syntax structures of like sentences and stuff. So from there, I kind of, I guess, got to understand how some of the, I guess some of the terminology that the people that are, that are working on the project, some of the terminology that they use.
Diane: Sarah is our second example of a student who got involved in research after having experience with just first-year programming in Python. Of course, her interest in, and knowledge of, linguistics also helped quite a bit!
Mario: With Sarah, we saw how a chat about courses lead to joining a research project a few months later.
Diane: That’s right. But sometimes, a connection you establish with a professor can benefit you later.
Arkaprava: So, back in first year, I was doing a seminar course, which was started by Professor Alan Borodin and, and there was a guest lecture by Professor Henry Yuen about quantum computing. … it was really inspiring. I got to learn a lot of new details about how quantum algorithms different from classical algorithms, how quantum computers work, I just remember my head was on the brink of explosion. (laughter)
Mario: That’s Arkaprava, a third-year student in 2022-23. And if you are thinking he later joined Professor Borodin or Professor Yuen’s group, you’d be mistaken. Arkaprava shared what he did after being enamored by quantum computing.
Arkaprava: So, I started reaching out to professors, starting with Professor Henry Yuen and other professors in the CS and Physics Department about quantum computing, saying, I'm interested in learning more about this and potentially doing research. And several cold emails followed.
Mario: To which you of course, you've received all the responses.
Arkaprava: Of course, yes, all of them were immediate responses within one or two minutes as well.
No, most of them were declined, of course, I was a first year I didn't really know much, to say the least. But then, as I said, I was in the seminar course, that being taught by Alan, and he got me in touch with Professor Nathan Wiebe, who had just joined UofT a few months ago. And so I contacted Nathan Wiebe as well, and I was able to join his informal study group with a few of his graduate students and learn more about quantum…
Diane: Arkaprava reached out to profs he didn’t know, starting with that guest speaker. Many students are nervous to do this, but it can lead to opportunities. In this case, Arkaprava was able to join Professor Wiebe’s reading group in his first year.
Mario: We will talk about cold emailing later in this episode, but for now let’s focus on Arkaprava’s story. And if you are thinking Arkaprava joined Professor Wiebe’s group, you’d be wrong again! Is there an evil witch laugh we can insert here? Yes? No?
Diane: I think Mario’s gone off the rails for a moment. Arkaprava did eventually work with Professor Weibe on a research project, but it wasn’t until later. Arkaprava shares how the story comes full circle.
Arkaprava: So, in my second year, I was doing a physics course, in which the lecturer mentioned that there was a scholarship opportunity provided by the Center for Quantum Information and Quantum Control, CQIQC at UofT.
Arkaprava: ...CQIQC takes in a select number of undergrads every summer to do some research with one of its members. And Nathan was one of its members. And I had some experience in algorithm I had some more experience in algorithm design by that time, and some more physics experience as well. And I could apply to that, as well. And I was able to demonstrate sufficient interest and then that I was able to secure the position.
Mario: So, it looks like there was around a year there between being part of a reading group and landing a scholarship from the CQIQC. But where did Arkaprava get that experience on quantum algorithm design?
Diane: This is one of those interesting situations where being part of the reading group helped Arkaprava land two opportunities. Come February, first-year Arts and Science students are eligible for the Research Opportunities Program (ROP), which is a way to do research for course credit. Arkaprava tells us more about how being part of the reading group, and work he did on his own to pursue his interests, helped him land that opportunity.
Arkaprava: In the list of topics I saw for the ROP, one topic was about implementing a quantum algorithm. And I had learned about this algorithm, and the study group which really helped me a lot. So I was able to speak about my knowledge of what the algorithm is, what it does, and how it's meant to be used and propose a few ideas for its implementation.
Arkaprava: Yes. But it wasn't just attending research groups, I was also trying to implement some other simpler algorithms on my own in my free time. And I could, I could refer to that as well in the cover letter.
Mario: So, to sum up that timeline, Arkaprava joined a reading group on Quantum computing in first year, but also applied for an ROP. During the ROP, Arkaprava was able to gain experience with physics and algorithm design. Then, he learned about a scholarship opportunity from the Centre for Quantum Information and Quantum Control. He applied and secured the scholarship. What an incredible journey.
Diane: Absolutely incredible. And I want to highlight a couple things here. First, remember that Arkaprava sent out many emails before being connected with Professor Weibe and joining the reading group. Second, before applying for the ROP, Arkaprava demonstrated even more interest in Quantum computing by trying to implement some of its algorithms on his own time.
Mario: Those are great points, Diane. Networking and talking to others are a great way to make connections, and that takes work on its own. But going beyond those connections, it really helps if you demonstrate an interest in what you’d like to pursue. Whether that be committing to a reading group or implementing algorithms, or both, or something completely different.
Diane: And of course, pursuing an interest on your own is not just a way to prove yourself. If it’s something you’re interested in, learning about it will be rewarding and fun, and it will help you know if it’s a topic you want to pursue further.
Diane: So far, we’ve heard about the journey of three students that pursued undergraduate research at the University of Toronto. Each had their own unique path to finding that opportunity. A great way for you to start is by talking to a professor of a course you were or are in. Professor Mariakakis shares his insights here.
Alex: But I find, you know, in the right situation, if you enjoy taking a class with a particular professor, you like the topic of that research, and, of course, the professor is doing research, you might feel comfortable in it, especially if you're vocal in the class, if you're asking questions, if you're actively participating, go up to that person after class and say, Hey, I really liked the content of what we're doing. Or I like today's lecture, I happen to be looking for research opportunities, are there ways to get involved.
Mario: Remember that talking to professors and being vocal in class can be useful in other ways. During Anatoly’s journey, being a consistent contributor in his MAT240 class lead to a reference letter.
Diane: And in Arkaprava’s journey, a conversation with his first-year foundations professor about his interest in Quantum computing lead to a connection with Professor Nathan Weibe. But it’s also okay to reach out to people you don’t know. Arkaprava did it, and Professor Mariakakis confirms that it’s welcome.
Alex: I had a few undergrads that reached out to me over email saying, Hey, Professor Mariakakis, I'm interested in your research. I looked at your website and I saw this project. I thought this was really interesting. And I was wondering if you had any positions available?
And, first of all, the fact that they reached out to me, I was like, very, you know, very touched. This is awesome. I'm a new faculty person. but I also really appreciate the fact that they took the time to look at what I do ahead of time, they weren't just sending an email to every single faculty person that they could find on the website, but they took the time to just look, even just look at a website and say, okay, this person does human computer interaction, they do machine learning.
And I mean, I think they really went above and beyond. And they also said, Hey, I saw this recent project, or this recent paper that they, that you were working on, it doesn't mean that they knew everything that was going on, they didn't need to understand every term. But the fact that I kind of had a sense of, this is the kind of work that they were interested in, that really helped me out in terms of identifying people.
Mario: Diane, can you tell when a student has included you in some kind of mass-email mail merge?
Diane: Yes! The message is usually not specific to me or my research. And professors talk to each other. When email is generic, it doesn’t leave a great impression.
Mario: Yeah, I’m only about three years in, and I can definitely see a pattern with the emails I’m receiving. It tells me a lot when the student puts a bit of effort in to customizing the email. Speaking of which, Alex had some thoughts on one particular pitfall to avoid.
Alex: …Faculty directories get very rarely updated. So five years ago, maybe I said that I'm interested in, maybe I did say, I'm interested in health research, and five years from now I'm gonna say, You know what, maybe I just want to do cool new interaction techniques. It's using the same kind of skills in computer science, but it might be a different problem. And if you say, Hey, I saw that you do work in health, and all my papers are from five years ago, and maybe I moved on from that, maybe I won't be interested. So I think just, I think it's a fantastic idea to look through directories, look through websites, just be very careful and make sure that your, that what you're identifying as interesting, really ties to what that person prioritizes or is doing in general.
Mario: Diane, I feel like this podcast is only scratching the surface in terms of the opportunities that are available. What would you say was one key take away?
Diane: There were so many themes interwoven in this podcast. I think an important one is to make connections and get to know your professors. We just heard of three great examples of why this is important. And Professor Mariakakis re-affirmed it in his own advice. If you make these connections, a door you never knew existed might open for you!
Mario: What a great way to wrap up this podcast, thanks Di-
Diane: Hey, hey, hold on! What’s your key take away?
Mario: Oof, thought I might escape that. Conclusions are hard… let me think. Okay, I think I have two. The first is that it is possible to do research out of first year. We actually saw that in all three journeys, with Anatoly and astronomy, Sarah and linguistics, and Arkaprava and quantum. But also, I’ll repeat what I just said: astronomy, linguistics, quantum. You can get research experience in just about any area. So, don’t limit yourself. Think about your own interests and see what kind of research opportunities you can find there.
Diane: Well said.
Mario: (mumbling) Yeah, it’s great when I can read from a script.
Diane: What was that?
Mario: Nothing! Let’s thank our guests today for joining us. Thank you to Anatoly, Sarah, and Arkaprava for sharing your journeys with us.
Diane: And thank you to Professor Mariakakis for sharing your anecdotes and advice, too.