This month I'm making time to learn new skills. Continuing with my interest in machine learning and data science has to be somewhat extracurricular, because my current research project doesn’t really call for it. Well, I’m doing “regression analysis” by fitting models to data, but I’m not really practicing any of the ML techniques that I learned about last August at the data science summer school I attended, e.g., classification (incidentally, S2DS is currently accepting applications for this year’s class).
To learn some new skills, I’ve been spending 1 hour per day with the book Statistics, Data Mining, and Machine Learning in Astronomy by Z. Ivezic, A. Connolly, J. VanderPlas, and A. Gray (note they use the Oxford comma - let that be a lesson to you). This book is also known as AstroML.
The book is divided into two parts, and I’ve decided to work through it by bouncing between them. I’m not sure it is the most efficient way to work through it, but I think it will keep my interest longer. I’ve worked through Chapters 1 & 2, and now I’m working through Chapter 6. After that, I'll got back to Chapter 3, and so on. A research partner of mine and I have some ideas that we want to use ML techniques for. Hopefully this book will teach us how to do that.
There is a website that accompanies the book, astroml.org, and it contains the code for the figures found within. We are working through those, trying to understand as best we can, so we can implement similar techniques in our research.
This project has two main goals for me. The first, as stated above, is that it is interesting and will lead to some interesting research. The second is that it will really help develop my professional skills. I’m still waiting to hear back from a couple positions that I applied to, but I’m also starting to diversify my applications. In February, I’ve applied to 11 positions in industry so far - that’s in addition to the 20+ academic positions that I applied to during the Nov-Jan period. The academic application period is slowing down now, so if I really want to come back to the U.S. anytime soon, I’ll need to diversify.
That’s what’s new with me,