Friday May 27th
Research Week 3, the Dream Team and You do You
- We continued our research, and my mentor was nice enough to write an entire blog post on a research website about our project and our progress, and even linked to our blogs! How cool!? I’m super thankful for how much he is truly invested in our success and our growth, and to just being a part of this community. It’s been fantastic!
- I’ve also been learning German. Well, not really. I’ve been learning German colloquialisms from my research team; the first one I have learned is “You do you”, which makes me smile because I hilariously translate it to how it would be said in my home country (which would totally be more passive aggressive, with an eyeroll to add some spice lol), which is very different from the way they (my German peers) use it.
- I spent the week learning and adjusting some code I’m working on, as well as a mathematical model in terms of a formal proof / definition. We also lucked out in that we found both a really cool paper and another one was given to us that will both be of great help for us to build on moving forward. So I’ve been making my way through those as well. I realized that I absolutely love doing this stuff; I could do it forever! How can I make this my career!?
- My mentor said that the result already makes a really great theoretical paper, which is amazing!! It honestly hasn’t felt like work, but I have spent entire days realizing that it was late at night and that I had spent most of the day working on research. I believe this week one morning at 3am I also had a 90s - 2000s latin rock music sing along to unwind.
- Finally, we had a group meeting! So I finally got to meet the other student with whom I am working, who is a graduate student (finishing up) with a lot of knowledge, but with so much humility and just a superbly fun disposition, and just is an awesomely fun person to work with! I really could not be happier! He described our group as “the dream team”, which was shocking and made me simultaneously really happy! We’re excited for our progress and the energy towards the project has been really amazing.
- I received a grant to attend Oakland virtually this year via a workshop my WiCyS mentor told me about, and received my ticket to attend the AWM Research Symposium after the Roots of Unity workshop, which I had received a spot for in the Arithmetic Geometry group and was granted a virtual ticket to attend GHC this year (which is great because I’m doing a butt load of Maths classes this Fall because I just want to do Pure Mathematics forever and live until I am 200 so I can just keep taking Pure Mathematics classes because they’re awesome!!). I am so stoked!!! One of the mathematicians I’ve been dying to meet forever is giving a topology talk on Covers at the AWM Research Symposium!!! How cool is that!?
- I also heard from my BlackComputeHer mentor, who is amazing, wrapped up a mentorship with my NCWIT mentor at Facebook research (who is seriously the coolest cat ever! He’s so level-headed but fun), and worked on a wrap-up blogpost with my co-organizer partner in crime from ICLR (smile). I burst out laughing because we sort of stumbled upon an idea that we are now dying to try, and played around with it last night. It will be executed!! She’s so creative that even though it was a long day, we sometimes get into fun mischief that makes the meeting time incredibly fun!
- I was also invited to four (and counting!) events already in the city I’m moving to (including at one place / institution I’ve wanted to attend forever), received awesome swag (!!) (I will make a post about this but there is a lot I can’t talk about specifically), attended a book club, a few talks, and had a conversation with a company about my plans after graduation and what I was thinking of next (ha). Honestly, I haven’t thought that much about it at all. My thoughts that day stopped at “I’ll have oatmeal for breakfast”. So that was interesting to think about, and to hear from the market that my skillset is very much in demand (which is pretty awesome). Honestly, as I had mentioned, I’m more interested in building solid relationships, and based on those relationships, I would make a choice, I think. I’m more interested in that and how interesting the problems are than the prestige of a company or whatever.
- I will say that the skills I pick up this summer will be very important in doing good work on my current research project; they’re very much complementary. I’m hoping to improve on these skills so I get better and have a better intuition for how to problem-solve in this domain.
Stuff from this week
- I went to a stats book club this week in which we went over Hypothesis testing, Pearson’s Chi-Squared test, Simple vs Composite Hypotheses, Student’s Distribution, Power of Test, ( ), Mendelian randomization vs Clinical (funnily enough, a childhood friend just did her disseratation on this; she just graduated from Oxford in epidemiology). Asymptotically correct test, Type I vs Type II.
- I went to a zero-knowledge book club where we learned about GKR, verifier and boolean hypercube domains, proving correctness, witness and verifier and role of interactive proof in the system (we don’t need one; we can get multiprover interactive proof, probabilistically checkable proof), and are going into cryptographic techniques that turn interactive proofs into snarks next, as well as information theoretic work, IOPs vs MIPs, transformation into snarks, snarks vs interactive proofs. It’s a pretty solid group but it’s my first time attending. I look forward to learning more.
- I attended a couple talks “at Stanford” again, including one by Fei-Fei on vision and models, and learned a bit about traditional cognitive models and the work of Irving Biederman, JJ Gibson, RSVP (Rapid Serial Visual Perception), Cognitive models in Geon Theory, Objects composed of geometric shapes, Brooks and Binford and Fischler and Elschlager (components made with springs and objects), moving from hand-designed models to hand-designed features but learned models, domain randomization (different permutations of a rendered simulated scene; often these renders come from artists), scene graph representation, the concept of object permanence embedded in the knowledge graph (super interesting!) and even included Plato’s Allegory of The Cave in her talk! It was also cool to hear about BDDL, a symbolic, domain-specific programming language made for tasks (which makes sense because at its core it’s compositional predicate logic). Who says programming languages research isn’t pertinent to robotics? :)
- I also attended and led a seminar, and my internet dropped out for two minutes, but the group was so awesome that I came back in seamlessly, and they were so patient and engaged in our conversation. So that’s awesome. Next week we have another session on protocols in-depth, but I am not the deputy, which I was this week, so I don’t have to go all-out and make a slideshow and stuff like I did this week. However, you know me..if given the chance…. :)
- I’m travelling next week! I have a bunch of people who have already spoken to me about catching up (or meeting because I’ve been speaking with them forever and they live out there); it’s already a bit overwhelming. I also need to possibly pack some extra podcast stuff because I might end up doing a podcast out there (hi sirens!). I’m going to try to do what I can while I am there, but I don’t know much about how things are going to pan out, and my work comes first.
- I also found out that next year I’m doing some crazy travel, which is kind of scary but also exciting! I have a few loose ends to tie up, including some research stuff to get done, so we’ll see how everything goes!
And that’s it
Written on May 27, 2022