Yum Char: Discussions over dim sums

Had lunch with Quark. Haven't seen him in a long time. He's only free this week.(Oh ya, Quark, if you are reading this, the site I was telling you about is www.modepass.com).

Talked. Mostly on computer graphics, computer vision, AI and machine learning. Even in the computer engineering field, there's so many different classifications. We had a lot of ideas for a lot of things. We just need to sleep less to get things done.
Also, Quark and I come from the same school of thought. At least I am not alone. One of the first thing we have to realise as grad candidates is the fact that we are... grad candidates. So we have to think like researcher and less about application. However, I do not believe in wasting my time on things that is not useful in life. Quark agrees with me. We're more of the bottom-up than top-down people. Cool. Well, he finished his Masters way back and he still thinks like this.

Anyways, tried the annotation joke about Web2.0 on him. He totally gets it! Lol. I tried it on my colleague a few days ago. Fell flat.

So good talking to someone who understands what I am doing now!

I tried with some of my colleagues but heck, they don't even have a strong grasp of telco knowledge. Wait. that is not true. The only idiots I've met are all from this company. Oh well enough of that.

Anyways one of the idea we had was how to convey mathematical theorems and algorithms in human language. The tv series NUMB3RS did that moderately well. (I have the hots for Prof Eppes but that is another story altogether).

Here is a site: http://www.weallusematheveryday.com/tools/waumed/home.htm, deriving the whole idea from NUMB3RS to teach maths. I think it's alright. But definitely worth a visit.


Season:
2: Episode 14 - Harvest -- Hidden Markov Model.

Markov Chain Links - predicting probability of certain states in the future from what we know of the present. (Ex, was the weather yesterday: sunny, cloudy or raining. therefore what's the probability that it will rain today?)

Hidden Markov Model - In the event when we are not able to observe the actual state but only the outcome that is connected to the state by a certain probability distribution. (extend from above example, we do not know if it rained yesterday but the sales of umbrellas for yesterday was high).