The question What is Data Science is rather well answered by our professors and does not seem to merit another blog by an eager student. I would rather use this opportunity to write on another topic that struck me only today and my excuse for deviating from the question prompt (i.e. the given topic) is that the higher aim of the blog contest is to hone our skills as data scientists and if that can be done (better) by deviating from topic, why not!
The topic on which I would like to request the indulgence of my dear readers is “Slow Learning”. It is a little strange sounding name for a blog as would be expected from a home grown blogger, but even more exasperating perhaps would be the fact that the idea is also a bit strange. Slow Learning is nothing by Learning Slowly, a concept that I first discovered for myself while wading painfully through that most seminal of all AI courses AI and ML by Prof Andrew Ng on Coursera, and then I fell totally in love with it. I had reached week 4 of the course, I think and this was a really tough part about neural networks and back propagation. I had breezed through the earlier three weeks, trusting that I had understood everything, but in week 4 I was truly stuck. I was pointless to go on just by searching programming assignment solutions in github and so difficult as it was I reluctantly went back to week one and started all over again, only this time I was determined not to proceed to the next topic unless I had rigorously questioned my understanding on the present topic. The result was magical. I derived a totally different level of satisfaction from each lecture, assignment and quiz. That experience really sold me on to the idea of learning slow but learning sure.
In slow learning, as opposed to speed learning where the intention is to learn something as fast as you can and aid that effort by numerous tricks and tips like speed reading, memorisation techniques etc, the idea is just the opposite. The idea is to take your time in coming to grips with a difficult topic, to pause, replay, and go over again slowly and deliberately till you get every word and every line and then to test that understanding by questioning. The core of the idea is to understand a concept thoroughly without the pressure of submission deadlines, and the thoroughness of understanding is tested by questioning it and being able to come up with answers. If you find you are not able to explain a part to yourself very clearly, go back to the text or dig up some additional material. This will definitely slow you down but when the aim is not to merely complete a course (and in the shortest possible time) but to learn, then this slow journey is all the more enjoyable. Once you get used to this idea you will find that when faced with a difficult problem you would be able to trust yourself to find the solution, to be able to reason out and synthesise a complex problem, breaking it down into smaller and simpler chunks till you are able to solve a bit, then build up on it successively till you crack the entire complex whole. You would think, ponder, scratch your head and rack your brains, let the minutes and hours tick by but would not google the answer.
So friends this is the simple idea that I wanted to share with you all. I know it sounds a bit idealistic and to be honest I am only able to do it because I am doing these courses for myself interest and not to change my work profile or get a pay hike, but I do want to encourage you to try it out at least once and share your experiences with me.
Tae care and Happy (slow) Learning!