Just keep doing | Week 4 | MPhil Advanced Computer Science at Cambridge

In the past week I realised what it takes to have a good output: don't overthink what to work on next.
-
I experienced the problem with choosing between working on two mandatory coursework. I was debating which one I should do first. When I was working on the first coursework, I was thinking that I should work on the other coursework and vice-versa. However, both were mandatory! So clearly the problem was with me. So, I changed my approach to choosing one task and stop thinking about the next one - because I can't work on them simultaneously anyway. I noticed at least +25% more output by stopping worrying about the following tasks.
-
On a different though - learning is a slow process, and I believe the secret to scaling learning is compounding. I started tracking my actual learning time (not including breaks, transport etc) and I did 55 hours in the previous week. If I put it in perspective and think about the material I covered, it seems little. So, to start seeing the result of learning, I believe that the secret is to keep compounding this time because if you can do 50h over 40 weeks a year, that becomes 2000hours - which is substantial.
-
In the previous week, I learn about parallel computing and programming on a GPU, which seems very powerful and make the computations very fast. Also, I started to wrap my head around the ML technique of Gaussian Processes.
-
For the next week, I will focus on three main tasks: submit the coursework at Image Processing, submit the coursework at Digital Signal Processing and start writing the personal statement for the PhD applications.

Видео Just keep doing | Week 4 | MPhil Advanced Computer Science at Cambridge канала Andrei Margeloiu