Finally, after many many days of suffering sleeplessness, I finished my Masters of Computing in NUS. It seems not really helpful except that I find out that I am not good at learning theoretical stuff. I prefer solving concrete problems. Luckily I wasn’t doing a Phd, otherwise I will die on the way to accomplish it.
I took an hour to finish the book describing Johnny Bunko, and began to wonder how to apply the following six rules:
- There is no plan.
- Think strengths, not weaknesses.
- It’s not about you.
- Persistence trumps talents.
- Make excellent mistakes.
- Leave an imprint.
I am a front-end developer with 6 years working experience, and this professional is fading fast. Recent graduates are plentiful and far cheaper. I can’t be like this for the next 6 years without achieving significant improvements.
It’s about theory in machine learning. I screwed this course in B grade.
I don’t have the mind for mathematical induction and lost my memory of probability knowledge.
This is a very new course, and the flow of the content is quite categorised from the text book, cut from somewhere and another. There are just too much content to learn.
One lab, three take-home assignments, and one big project in group of at most two.
Prepare for hard-core math theories if you want.
There are two same modules in IS5152, I took the one from MComp. The course is more like a machine learning course in the sense of doing practical math problems, such as maximise utility functions, making optimised decisions (Operational Engineering). And there will be 5 questions in the final exam.
Two group-take-home assignments plus one final project in any selected research area related to decision making or machine learning.
There are many past year papers, and good math is necessary to ace it.