Chris is a first year PhD student and former CDS MS student. We talk with Chris about his experience in both the MS and PhD programs, life in NYC, finding housing, CDS culture and his research.
December 9, 2019 | 24:34
Tim Baker (host) [0:00] Welcome to the Center for Data Science Admissions Podcast. I’m your host, Tim Baker. And this week we’ll be talking to first year PhD student, Chris Ick. Chris is also a former CDS master student.
Chris Ick [0:28] So, my name is Chris Ick. I am a first year PhD student, but I also did my first year of the master’s program before switching from the master’s program into the PhD program. So a little bit weird situation there, but we can talk about that.
Tim Baker [0:28] Okay, so I mean, why not jump into it? Because people actually asked that question a lot. Can I apply for the masters and then transfer into the PhD program? So if you want to tell us how you went about doing that, that would be awesome.
Chris Ick [0:42] Yeah, of course. So basically, my background is actually in physics. I did my undergraduate at NYU, studying physics. And then my initial plan was to take a gap year, do some research and then go on and then apply for PhD programs in physics. Because I was interested in doing more research, I was interested in doing more physics and stuff like that. But over my gap year, I ended up doing research with Professor David Hogg, who’s a professor at CDS of data science and physics. And he got me more interested in the data science side of things. So I started flirting with the idea of doing data science, but I wasn’t sure if it’s something I wanted to do for five years. And because of that, I applied for the master’s program rather than the PhD program while I was simultaneously applying for a PhD programs in physics.. And then I spent the whole first year doing the master’s program in data science. And I found out that I really, really liked it. I really liked all the resources that CDS had to offer. I really liked a lot of courses we were taking and a lot of the material and the research opportunity, but I wanted more time to dig into more diverse research options. So one thing I became interested in was doing research and music and audio with Professor Brian McPhee, who was a new professor at the time, and I figured I’d have a better opportunity to do that in the PhD program than I would in the master’s program. So midway through the master’s program, I applied into the PhD program and and here I am today.
Tim Baker [1:59] So you had to do an application for the PhD program?
Chris Ick [2:02] That’s correct. Second application. Yes, I applied for my for CDS two years in a row.
Tim Baker [2:06] So how did you find the application process? Like if you had advice for people that are doing their application now, what advice would you give them?
Chris Ick [2:15] Yeah, this is something I’ve been thinking about a lot. Because we’re in sort of the peak of application season. I’ve been speaking to a lot of interested applicants, both for the PhD and master’s program. So going through the main components, right, your your statement of purpose, your letters of recommendation, and your GRE scores, right, because those are the things you can probably change now. Like, obviously, in your transcript, but it’s probably too early to change that at this point. So on those three things, I think it’s important to make sure you pick people who know you very well for your letters of recommendation and people with diverse perspectives of you. So I was lucky enough to have a lot of research background coming into this but I made sure that only to my letters were specifically research-oriented. And I picked a third professor who is more related to the field I was interested in which was music and audio and stuff like that. And then I had him write a letter about how I was as a student, taking his courses. For the statement of purpose – a lot of people want to talk a lot about their technical requirements and their research background and stuff like that. And I think that’s fine. But there are other places in the application for that to shine through. Whereas the statement of purpose is the only place for you to really talk about you as a person, your motivations and the sort of reasons you have for pursuing a graduate degree in data science. So I advise you spend at least half if not more of your statement of purpose talking about you and why you want to do the program, what you bring to the program, etc, etc. More of the sort of softer skills and things about you that may not come through in like a formal research list of research background or a resume or a CV or something like that. And finally, your GRE scores. I don’t know best second thing for that is don’t stress about it too much. You could take it every two weeks. So I would start by taking it right away as soon as you can and getting your assessment on how much more you need to study before you can get a score you’re satisfied with because I know a lot of people will spend I don’t know 10 hours 20, even 15 hours a week studying for the GRE for a month, two months in advance, which is totally fine, I’m sure do very well doing that. But that’s energy and resources that might be better spent focusing on your statement of purpose or somebody like that. So try to get an assessment of how much more you need to study for the GRE before really stressing about that, because people seem to put a lot of emphasis on it in terms of the amount of work they put into it, but it’s only one part of your application.
Tim Baker [4:24] I mean, you have a unique situation, because you were here already. But how did you find building a relationship with faculty? Did that help with your application?
Chris Ick [4:34] Yeah, so for my scenario, specifically, I was applying to work with Brian McFee because I have no background in audio/music but I was interested. And what I did was I sent a cold email, which is the standard like I do some research on Brian McFee. I looked up some of his papers and looked at his lab website – all of the faculties lab websites can be accessed from the CDS website. So that’s a good place to start. And then I would try to find professors who would have the work that you find interesting or inspiring, and try to get background to see whether or not it’s something you’d want to work on for the next three to four years if you’re interested in working in a PhD program. And for the most part, if the professor isn’t a super famous or super busy professor that will always be happy to respond to you and talk about it. And some of them even more willing to sit down and schedule office hours with me. Okay, so for me, I emailed Brian McFee saying, I think audio is pretty cool. I think your papers are pretty interesting. I’d love to sit down with you and chat. And then a week later, I sent the follow up email because I didn’t get a response because he’s a very busy professor. But that follow up was enough because he said, ‘yes, let’s sit down’. I set up a time of day to sit down and chat with him. I started going to his group meetings, I started going to a class that he recommended and stuff like that. And I started building a relationship with not only him but the field. And I think that was a really good way to sort of integrate myself into his field because it became more concrete that I was someone who is willing to work in this field and build a background and started working in active projects there. So start with a cold email and then work out from there is the best advice I would say.
Tim Baker [5:55] Excellent. So let’s talk about actual life at NYU. You’ve been at NYU for a while. When you started in undergrad, did you? Did you already live in New York?
Chris Ick [6:06] No. So I’m from New Jersey. Stone’s Throw which is only an hour away, which is nice. It’s far enough that I feel away from home, but close enough that I can go when I want. But I did it cold turkey – I made sure for the entire first semester, I didn’t go home at all. Because I knew if I started going back and forth, that I would never really integrate. And obviously not everyone has an option so the cold turkey might be a good option in general. Um, the way I think about NYU is it’s a huge University, right, 23,000 undergraduates and a similar number of graduate students, I believe. But it’s obviously there’s no huge large community that like all go to the sport games together or stuff like that (sports at NYU isn’t really that big.) But I do find smaller, more specific sub-communities that I really get along with. So off the top of my head… tonight, I’m going to a very informal talk series in the physics department because I met a lot of people there while I was doing my gap, your research and I know all are interested in the sort of casual, almost silly talks that we do every Friday. And that’s the one sort of circle of friends I have. I have a circle of friends in the master’s program. We all just met during orientation, actually. And we went out for drinks together, and I still hang out with them and see them regularly. In fact, I was just chatting with them before coming over here.
Tim Baker [7:18] I actually was on my way out, and I overheard the conversation.
Chris Ick [7:22] And I really like video games. I really liked this video game company called Super Smash Brothers Melee. And there’s actually a club for it at NYU. So I regularly go there and compete in the tournaments and meet people that way, and they have a separate circle of friends outside of my academic friends. Yeah, it’s a really good way to get like sort of diverse backgrounds of people because then that way, I’m not always hanging out with just data science people or just physics people and stuff like that. And of course, like you can intermix those groups of people and find your community that way.
Tim Baker [7:50] So coming from the MS program into the PhD program, how did you find that transition? Like what’s different about coming from a traditional ms program and then going into like the research program for a PhD?
Chris Ick [8:04] It’s a bit tricky for me to say because while I was in the MS program, I was taking it part time. So I was only taking two classes a semester, and I was working as a research assistant. So because of that, I went from that to taking three classes a semester, but one of those classes is a research rotation. So it’s pretty similar. The biggest difference for me is obviously having an office here, which it’s nice having my own space I can work in. Academically, the course-load is very similar for the first two years or so. So the coursework hasn’t changed that much. But I do find that the support I get from my staff and events and stuff like that is definitely more oriented towards research and fellowships and scholarships, and, you know, typically more academically minded environment. So one of the reasons why I really liked the MS program, but I decided it wasn’t a good fit for me was there’s tons and tons, almost too many resources for job placement, the hiring and stuff like that, and I I thought it was really interesting. But once I determined I want to do research, I thought I should steer my sort of energy towards getting research roles and fellowships and stuff like that. So swapping over sort of changed the mindset of the people I work with, as well as the sort of resources that are targeted towards me to be less workforce oriented, more research oriented, if that makes sense.
Tim Baker [9:20] Yeah definitely. This is probably something that you’re old hat at, but finding housing in New York City… any advice that you can give to people?
Chris Ick [9:29] So from undergraduate to today, I lived on NYU housing for two years (as an undergraduate it’s provided) but after that, I have been bouncing around apartments. So I’ve gotten really, really good at finding apartments, partially because you have to in New York City, and partially because I’m very, very picky. In the past 18 months I’ve lived in six different apartments. And I would not recommend that by the way, that’s that’s me being really really picky and me being in living situations that were ok but I thought I could do better so I bounced around a lot. But finding housing in New York, it seems really intimidating at first, but the more you do it, the easier it gets. And if you take action and move fast, it’s not that bad. And so my usual strategy – there’s a lot of really good communities online for finding housing. Facebook is a really, really good network because it’s one of the most popular social networks, obviously. And there’s groups specifically for NYU housing – there’s NYU housing subletting and renting, NYU housing, and there’s all sorts of non-NYU affiliated ones like Gypsy housing, Ghostlight Housing and there’s a bunch of NYC housing based things. There’s also Craigslist, right, which a lot of people are skeptical about because you know, you always hear about scams and stuff on Craigslist, but in my experience, the best housing situations I found, have also been on Craigslist. So in terms of strategy, I would say, go online, go on Facebook, on Craigslist, do as much research as you can research.. Reach out to as many people as you can, but don’t spend too much energy trying to like figure everything about the housing online because you’ll just exhaust yourself that way. Because people are bad at communicating online and stuff like that. What you want to do is just try to get your foot in the door to as many apartments as you can to see them in person. So obviously never rent an apartment that you haven’t set foot in and been inside of. It seems obvious to you and I, because you know, we’ve been here so long, but you always want to see the apartment before you ever put money down on it 100% of the time. But once you’re there, and you have the person or the realtor in front of you, then you can ask all your questions like how much is rent, how long is the lease, what utilities aren’t, aren’t included, etc, etc. So for me, I think this is this is more personal, but I find that location is much more important than the quality of housing, because, for me, I spend most of my time working or out and about enjoying New York City. So I spent a lot of time basically just picking one or two neighborhoods, and then tried to schedule as many viewings as I could in that neighborhood in a day. And over your many years at NYU, you’ll like develop a network of friends and co-workers and colleagues and stuff like that and that will also be a really, really good way of finding housing. So I was bouncing around like Craigslist and random housing places for a while because I wasn’t really satisfied. And then one of my old colleagues from MIT Physics Department reached out to me saying ‘hey, I’m in need of a new roommate, and I heard you were thinking about moving again’… And that’s the best because then you avoid a realtor fee. And then you know, you’re not moving in with a crazy person, etc, etc. So that’s the best option. But that’ll come with time. And until then, just being persistent and putting in the time and effort of seeing apartments in person is the best way to find housing.
Tim Baker [12:39] And so with regards to CDS, we’ve talked a lot about the academic side.. How do you find the social side like the general social culture of CDS?
Chris Ick [12:51] So I’d say it’s analogous to what I said earlier about NYU. So the master’s program is very large. There’s I think, 150 students in each class, right now? Something close to that. But within those classes I find smaller sub-communities with common interests, common goals, common mentalities, etc, etc. So like I mentioned before I have my group of friends I met in orientation – it’s just a group of like minded people who all happen to get along with each other relatively well. So we share, we have a group chat that we all talk about stuff in, and we talk about, like social events and stuff like that. Um, and then the PhD students, I find it much closer because it’s a much smaller class. It’s only I think my class is 13 students or something like that, right? And there’s only three classes right now. So that’s a very small group of people who we can all hang out and chat with. I think it’s 32 or 33 people total.
Tim Baker [13:40] It’s like 31.
Chris Ick [13:41] Yeah, 31. That’s a very manageable amount of people. And you know, everyone has their own lives. Some people are more present than others. But I find that whenever I want to chat with someone, either about work or about the homework we’re doing or about hanging out. It’s very easy just to walk into someone’s office and start a conversation. So for example tonight, that talk series I mentioned earlier – I invited a couple of people from the PhD program to do that with me and everyone seemed really interested. And again, there’s obviously group chats and online interaction and stuff like that, that make it much easier.
Tim Baker [14:08] Okay. So how do you find your like work/life balance? I mean, it sounds like you find a pretty, pretty good right now..
Chris Ick [14:16] Yeah. So it depends on how much you want to how much time you want to spend in the office and how much time you want to spend at home. So one nice thing about working in data science is that since all of your work is on a computer, or in a cluster somewhere, you can work from home basically whenever you want it within reason. But for me, again, since most of my social life exists in the department around people in the department, I like coming in. I like spending time with people I like talking about my problems over a coffee break with someone in my class or something like that. It is what you make of it is what I’ll say. So I personally prefer having a lot of work. And I like spending a lot of time on work because I find that my social time and my interactions come out of that. So again, that’s a function of me taking a full course load and also teaching this semester so I do spend I’d say about 50 hours a week working, which is pretty good for a PhD program, honestly. And that’s spread out too. So for me, that means working a bit on the weekends, so I can get home at more reasonable hours during the week and stuff like that. But I know people who don’t work at all on the weekends, and they get everything done. I know people who will only work between 10 and four, but then will work most weekends as well. So you can balance it however you want as there’s a lot of flexibility being in the PhD program, which is really, really nice. I never feel like I don’t have enough time for myself, right. I always have time to cook dinner at the end of the night, I always time to like, sit down and chat with my roommate. I always find time to do things that I want to do.
Tim Baker [15:42] So another question I get a lot from potential applicants is what kind of research is going on here? Are you comfortable talking about your research?
Chris Ick [15:51] I would love to talk about my research. In terms of all the research at CDS…
Tim Baker [15:56] Well, let’s talk specifically about you.
Chrisk Ick [15:59] Yeah, I don’t think there’s anyone in this department who can talk about all the research of CDS, which is, you know, great. That’s the reason why I actually picked CDS was because it gave me the widest breadth of research I can explore. But I’m working on two separate projects right now. So I’m still dipping my toes in physics. I’m working on a project on analyzing solar flare data using a Bayesian method called gashing processes. So we’re modeling noise level phenomena using a statistical method. And the reason why this is fun and interesting is because the solar physics community hasn’t really graduated past ori analysis, which was developed, I’d say it was really in its prime in like, I don’t know, like the 50s and 60s and stuff like that. So there’s a lot of opportunity for improved mathematical methods on existing work. So we’re proving that right now, which is really, really exciting. And I think I’m gonna have a publication submitted either by the end of this year or early next year. Depends on how things go. But obviously, and that’s with Professor David Hogg. I’ve been working on that for a while now. But it’s a great project is a lot of fun. So that data set is one dimensional time series data. It’s just brightness as a function of time. And I became interested in what other datasets had the same form of one dimensional time series. And audio was the first thing that popped out. So I go to a lot of live music in New York City. That’s one of the nice things about the city is having access to a lot of like cheap concerts that are very good. I like music, I play music, I produce a little bit of music. So I thought music would be a cool thing to sort of explore using the mathematical and technical tools developed in physics. So I reached out to Brian McFee, and we got started a couple of projects. So I’ve been working on a couple of different things with him. mostly small side projects. I use Gaussian processes, again, to do tempo analysis. So we can get a probabilistic estimate of tempo, which, as far as we know, hasn’t really been done before. But it’s also hard to motivate like a good reason why that’s better than current methods. So we didn’t really we kind of played with that project for a while. I did a lot of stuff on speech analysis for a large collaboration that Brian was working on. I was working on tools for speak source separation, speaker diarization – so if you and I having a conversation and we look at this audio file, can we automate who’s speaking at what time and then silence detection. So those are all for the same project. I did some research over at Amazon over the summer when I was interning, which was on recommender systems and smart music recommendation. So like analog see, like discover weekly on Spotify, or other smart music curation tools. And then right now I’m working on a project on urban audio. So my project right now is coming up with a way to basically come up with ways to basically assign spatial labels to audio so given one microphone, two microphones or some configuration of microphones, can I figure out where the audio sources coming from? So this is the problem that solved for fixed configurations. But we want to generalize it out to any set of microphones and only configure spatial configuration. And this is a lot of benefits in terms of mapping and tracking urban sound and environments like New York, right, which is part of a larger project that our lab has been working on for several years. Yeah, it’s a fun project. And it’s nice and able to do two different, very different field. Yeah, at the same time, and still find a lot of parallels between them. And that’s something CDS actually encourages us to do via the research rotations, where I believe one or two of them are required to be outside of your proposed field. Yeah, so no matter what you do, you always end up sending something out of field, which gives you a lot of opportunity to explore things that you might not have thought about before.
Tim Baker [19:27]
Yeah, that’s awesome. So you’ve mentioned that you’re working with David and Brian, and how is the support from them when you’re doing your research like you have constant access to them and, you know, sit downs and stuff like that?
Chris Ick [19:42]
So I like having both of them as my advisors because they both give me very, very different styles of advising. So that’s nice for me, because I can I can kind of sample different types of advisors, advisors and see what I want as my parent advisor. So David Hogg is much more laissez faire I would say. He very much believes in my own independence and my own ability to perform research and do stuff on my own, which is nice because that means I can work on a project on my own for a few weeks at a time or a month without him like micromanaging me or getting too involved. And it allows me to work very independently and take my own paths in the research, which is something I really appreciate about working with him specifically. Brian is much more hands on; we have weekly sit downs, where we talk about research, we talk about fellowship applications, we talk about classes, we talk just about, like how I’m doing in terms of like, the program like life situation, etc, etc, which I really appreciate. And Brian is super available all the time. We work in this we work in the same building. So it’s very easy for me to get a hold of him whenever I need him. And that way, he’s usually pretty on top of what’s going on. He knows what I’m working on specifically. So yeah, two very different styles of advising and two different students might prefer one or the other. For me, I find that I like working with Brian a bit more because I like having like, I like sort of bouncing ideas off of him and Like, doing sanity checks to them to make sure I’m doing things properly. But you know, to each have their own.
Tim Baker [21:06] Okay. Excellent. And then just a couple quick wrap up questions here. Do you have any sort of final advice for an applicant, just something you touch upon that you wish that you knew before you applied?
Chris Ick [21:22] One thing I meant one thing a lot of people have been talking about is how do I know what I want to do? How do I know if I want to do a master’s program? I don’t know if I want to do a PhD program. How do I know if I even want to go to graduate school? And if so, where do I go, etc, etc. And if you’re interested in data science, you’re probably micro optimizing for every possible decision factor there could be whether it be like the location of it, the program, the cost, the income you’ll make after taking the program, etc, etc. And I’d say, you know, if you’re in this mindset, if you’re looking to programs and data science, you probably have a successful career ahead of you, no matter what you end up doing. But in this moment in your career, when you’re picking a graduate program, you have a lot of flexibility. You can really do whatever you want to some degree, and research wherever you want and study whatever you want. And you should take advantage of that flexibility. So when it comes down to making the decisions about what kind of program do I want to do, where do I want to go, etc, etc, I would just try to think about what feels right and what you think will make you happy, short term, right? Because again, like, with a strong enough background to do data science at NYU, you probably have a successful career regardless of whether or not you get the degree. So you should focus on doing things that you enjoy, and you find intrinsically engaging. Because if you just want to make money or you just want to get a good job, you probably already can do that. So yeah, I guess I think we think with your heart for a bit, which is hard for a lot of like, technically minded stem people, and it’s something that I still struggle with myself, but, you know, I kind of took the plunge and I applied to this program, and I’m happy that I did and don’t regret it at all so, you know, take a risk and just go for it.
Tim Baker [22:55] That’s perfect advice, because I get that I get the question, what should I do? I don’t know. That’s a decision you have to make, man so follow your heart and get there.
Chris Ick [23:07] Yeah, just do it.
Tim Baker [23:08] So last question. What is the one must do in New York?
Chris Ick [23:14] Oh, man. That’s a tough one. The first thing I think about is food, obviously. So follow my advice. Okay, if you like spicy food, Dan and John’s Wings in the East Village on First and St. Marks has some of the best buffalo wings I’ve ever had. And they’re also incredibly spicy. And I’m talking like probably shouldn’t have them if you’re not comfortable with spicy food because you’ll get sick or something like that. But I think about those wings on a regular basis. I try not to go too often because it’s not the healthiest food. But yeah, if you like spicy food, that’s the place to go.
Tim Baker [23:54] Excellent. Cool. Thank you so much. Yeah, of course. Thanks, man. I appreciate it.
Tim Baker [24:02] Thank you for listening to the Center for Data Science Admissions podcast. If you have any questions regarding the admission process, please email us at datascience-group@nyu.edu. The music for this podcast was composed by the instrumental artist Cryptic One. You can find his work on his bandcamp profile.