Method of images, BBN, NYU fellowship, Entrepreneurship homework and Machine learning astrophysics grant
Today as usual was exceptionally busy yet again, let’s get right to this post!
Mathematical methods tutoring
Tutored today as I usually do every Tuesday morning at Fisk University for students taking mathematical methods for physicists. I often times think as I’m fielding their random math and physics questions how grateful I am to my past self for always taking the extra courses. It’s something I’m honestly very proud of, maybe that which I am most proud of in myself; always making the choice to tackle the hard problems and learning as much as I could. It’s the reason I learned Linux in highschool, the reason I know more about differential geometry than the typical physicist and why I know what I do about stellar evolution and exoplanets though I wrote my dissertation on string theory and inflation. AND, why I know about fluid dynamics and turbulence. It was hard, and continues to be trying when entering a completely new field and being confused about some topic, but it’s nice to look back every now and then and just be happy to know I know what I know!
About tutoring– so I was asked about a method of images problem. Basically, here’s my morning: get up, shower, eat some breakfast, do some short exercise and get in my car and head to Fisk. Park my car, walk into the library and BAM method of images, and BAM answer the question. If you’re not familiar with physics, method of images is not exactly difficult, but I wonder how many people 1.5 years out of their PHD and 5 years removed from coursework still recall that random technique..?
Anyway we went on to solving general second order differential equations and then did have to resort to the book haha :P.
Ted talk update
Today I also completed my Ted talk slides since tomorrow I’m giving a practice talk to the organizers– I hope they like it!
Bob and I did a pretty thorough investigation to verify that no one’s written a paper on an idea we are currently working on. No one’s worked on it, so it’s a go! Time for some calculations :). I’m looking forward to this part of the project since I’ll be getting back into some analytical calculations. I usually do most of my work coding either some analysis scripts or some computational model, but it’s nice every once in a while to get back to the math of it all. Even though computers do the heavy lifting, I see true value in analyzing toy problems to get an intuitive appreciation for a process.
Ok, I didn’t interview for a position with facebook myself, but my good friend did :). He’s currently a data scientist at intel and someone I studied math and physics with at FIU. He’s one of the most intelligent people I know, I used to call him “he who knows all math” when we were students.. and I think it’s pretty true still today.
Anyway, just as a test, I asked him to pretend he was the interviewer and to ask me the questions (see if I could make it!). Well, I’m glad to see I was able to answer everything with some Pandas hacking, but he said that the interviewer requested the explicit use of SQL. That took me a bit longer, to remember the difference between inner, left, outer joins, etc.
The interesting problem was: “what approach would you take to design an algorithm that figures out a user’s birthday?” I’ll leave that problem to the reader :) but I will give you some hints: think about many languages, not just English, and also remember there are other big life events people have such as anniversaries and graduations, promotions, etc.! Not easy…!!!
Well I’m extremely excited about a fellowship I’m applying for at NYU. They have a program that is a two year postdoc that leads to a tenure-track faculty position. I usually wouldn’t be so excited about a faculty position I think, except it’s freaking NYU. They do exactly what I am doing and want to be doing– data science applied to astrophysics. I can’t imagine doing research alongside David Hogg and Yann LeCun. I think being a career researcher in machine learning and deep learning is such a dream. It’s really the type of position I’m trying to find, whether it be in industry or academia. Let’s see..
I’m currently putting together a grant proposing the use of data sciece methods in astrophysics problems. I don’t mind doing it, it’s nice writing papers or grants to help solidify your knowledge on something– but I am under such tight time constraints these days, it’s hard to do it all! Which reminds me, I need to complete my AAAS Policy fellowship application..! Oh well, at least I know this grant is going into a paper my colleague and I are putting together.
And, if that weren’t enough, I also completed a homework assingment for the Wondr’y Entrepreneurship program I’m in (see previous blog post about that!). My business plan is essentially a data science consultancy service assisting in data-driven business decisions. Should work out alright I hope!