All of this post is just going to be the productivity check in, buckle up.
I have a really hard time finding time to read. A part of this is because I’m very vulnerable to short form content. Instagram Reels/Tiktok have the unique ability to waste hours of my time without me realizing it. The solution isn’t just to uninstall the app. It connects me to my friends in a way that I value, and uninstalling them leaves me feeling isolated. Another part of this is that I’m just lazy. I find that I don’t work as intensely as I want to, and it’s really hard to get me to start working once I stop. I spend a lot of time doing things that are “productive”, but aren’t pushing me towards a major goal.
One solution is to try and get most of my work done in small section of time. This follows the pareto principle, complete 80% of the work, in 20% of the time. However, a lot of common wisdom (and much research) says consistency is key to achieving long term goals/sticking to habits. However, I’ve never been able to pick a certain time of day that I can devote to just one task. In my experience my life right now is just too chaotic. In addition to that, many of the tasks I want to complete don’t need to be habits and many of my goals currently are short term. So I think I will adopt this burst method approach.
My plan is to work in 45 minute sprints, where I’m focused on reading as much of a book as possible (while still comprehending) or finishing as much of a project as possible. I’ll try and tackle one weekly task in one day. At the end of 45 minutes I’ll take a 15 minute break, or if I’m so in the zone I might just push the break till the end of 90 minutes and take a 30 minute break. We’ll document progress and see how it goes.
I want to get to the exciting stuff like new advances in deep learning and nuclear fusion, but I keep running into walls! With deep learning it’s easier, I can kind of learn on the fly and bullshit my way through things like you saw me do with the last post. But in other fields where I have even less of a background this isn’t going to fly. If I want to be able to make contributions in many of the fields I’m interested in, I’ll need a significant amount of prerequisites. The first three being multivariable calculus, linear algebra, and statistics.
I read a really interesting article about how to study topics with a large amount of dependencies by studying multiple of them at the same time. The basic idea is that you’re deliberately slowing down your learning process to give yourself time to understand the material. Lets say each subject takes a month to study if you only focus on that subject. Instead of learning each subject a month, you learn at 1/3 the pace, for each subject, just all at the same time. You can read the article for more of the scholarly support, but I think intuitively, it makes sense. Many researchers I look up to also recommend it.
This also has the added benefit of introducing me to all the dependencies I need (and how they interact) all at the same time. I think I’ll be taking a step back from reading papers and trying to contribute. I’ll focus more on getting my foundations in check.
I’ll let you know how it goes.