آموزش باید از مدل انتقال دانش از استاد به کاسه ذهن دانشجو به مدلی بازتر که در آن استاد عموما علاقه به کسب دانش ایجاد میکند تغییر کند.
نقل به مضمون از کارو لوکس، کلاس درس منطق فازی

]]> http://mahdi.milanifard.com/blog/?feed=rss2&p=2038 2 http://mahdi.milanifard.com/blog/?p=1998 http://mahdi.milanifard.com/blog/?p=1998#comments Sun, 20 Jun 2010 03:12:49 +0000 Mahdi http://mahdi.milanifard.com/blog/?p=1998 This is from the Google Analytics report for the past month on my blog. I had noticed that recently Google Chrome is begin used more often among my blog viewers. But this month it has passed Internet Explorer, now taking the second place after Firefox:

My audience, however, is largely biased. Looking at the worldwide market-share here, as of May 2010, IE is leading the game, with Firefox and Chrome following it with a rather large gap.

At any rate, regardless of whether it’s the second or the third, Chrome is definitely gaining ground in the browser market.

Carl Sagan – Pale Blue Dot

This is part of chapter 1 from the his book “Pale Blue Dot”. You can listen to the audio book here. Found this on a friend’s favorite quotations.

My mom was a primary school teacher. We, the kids, had to study hard all through our education to please her. I always had this feeling that she did not like to see 19.75′s out of 20, the highest you can get bellow a full mark on the Iranian marking system. She was not even content with us being the top student in the class, or the school for that matter. She wanted full marks, period. But don’t get me wrong, she was pushing too hard or anything. She was just not pleased to see us less than perfect.

Well, I did score a perfect 20 all the way through the primary school. But this could not go on forever. Maybe because I entered a special training program with math-intensive courses. These were the best of city of 5 millions. I really had to study hard at some point to keep close to the top of the class. I remember that it more or less hurt not to be perfect, coming from the mentality of the primary school. All in all, high school was not fun, at least not the long hours spent on courses with lots of points to memorize.

The first year of undergrad was a complete turning point for me when it comes to academic life. I had moved to another city and there was no mom on cite to please. The outside pressure was more or less gone and I had my own self to please this time. So I started slacking, but in a good way.

Anyone reading this (yes, I’m talking to the two of you insomniacs that have nothing else to do) would agree that one can get fairly close to his/her optimal performance without putting much effort. The last few points out of a hundred, however, are quite tough to get. What I began to realize was that I could easily get a near perfect grade with only understanding the fundamentals of the material in the course, and a few hours of light study before the exam. The result was amazing. I had plenty of extra time to spend on my other interests and I was the top student in our major. The interesting part was that, at least in the first year of undergrad, I did not score the best in any of the courses I took. I was, however, one of the good ones in all of the courses, making my average the highest.

There is a lesson in this for all of you who are still struggling with your studies. What comes next are my suggestions, as a veteran of many classes and exams, on how to build up your own effective study program:

• Be reasonable. Don’t expect anything from yourself more than “doing your best”. The result is only secondary to the effort.
• Don’t try to do your absolute best. It will fill up your life with stress. Stress is the killer. Get it out of your system.
• Settle to be reasonably close to your best performance while minimizing the effort. You need to find your own way of doing that. For me, it was doing the homework and leaving the classes with almost no questions left in my mind.
• You don’t need to be the best in anything specific to be the best in your field. Keep that in mind for undergraduate studies.
• Stick to the general ideas and fundamental points of the course. Make sure you are comfortable with them. Details are secondary. You can cram them before the exam.
• Try explaining the material to someone that doesn’t know much about it, like a fellow classmate who has missed a couple of sessions. This has mutual benefits for both sides. It helps you get a better grasp of the general ideas.
• Always, and I mean always, reserve some time for something that you enjoy doing. This includes the night of your exams when you are behind in your studies. Of course you need to be smart about it. As for me, I would reward myself for some measure of work I had done, say 10 minutes of game play for 10 pages of notes that I would study.
• Last but not least, experiment with different studying schemes. Don’t stick to anything that causes stress. And remember, after all you are supposed to enjoy your life, one way or another.

A long post indeed. Hope it helps some fellow students have a better time with academia. I would appreciate comments on this, but please under this blog post.

The year is 2010. Planes are very big. They can take 1400 passengers… Passengers wait in departure boxes. Robots load the boxes and put them onto the plane… There is no pilot. There is no crew… Robots are flying the plane…

This was part of an article on the future of flights written for an English course in the early 90′s. The pictures, as far as I can remember, included metal-faced humanoid robots doing the jobs of the crew. This, of course, looks rather ridiculous today. But if you look back at the beginning of the modern computer era, you will see such dream-like expectations all over the place. Computers taking over the world and apocalyptic wars being waged on human beings are elements of many sci-fi movies. Yet I can tell you, confidently, that no such thing is about to happen in at least half a century.

As a researcher in the field, there is an interesting question for me when it comes to the dreamland of super-human machine intelligence. Why aren’t we there yet? Was it literally unachivable, by the 2010 deadline, to build humanoids to replace a crew?

While there are many reasons as to why we failed to achieve such expectations, I would like to talk about the less obvious, and possibly the most influential reason. It was not economically justified. It is also not economically sound to build an army of deadly flying robots with a central AI control that might some day turn against human beings. It is as it is for now. Maybe someday one would spend a million dollars to replace a flight attendant that would cost a fraction of that over the lifetime of a robot. By that time the oil prices are high enough to prevent most of us from experiences the trill of such flights.

I am not, in anyway, suggesting that we have developed the level of intelligence needed for tasks like taking over the world. What I’m suggesting, instead, is that we did not move in that direction because the money was going in another way. Auto pilots, for instance, are way better substitutions for pilots than humanoids sitting in front of flight controls. Robots are appearing these days as clean and well-behaved replacements for pets rather than machines to carry “departure boxes”. Cheap labor, of course, would take care of that now. As for the plane, the latest Boeing 787 “Dreamliner” (to be in commercial use in 2010) is designed for fuel efficiency and not for thousands of passengers.

It seems rather sad to think of science as yet another product of the economy. But in today’s world everything is more or less governed by the money. The flow of science and its direction is no exception. So for you reading this, I can assure you that unless some crazily rich mad-man or government is going to sponsor a “terminator” project, no such thing is going to be built. And if you are a crazily rich mad-man, please don’t do it, and also give me some money.

Let me finish this by quoting the ending of the childish article on the future of flights:

[A robot:] It’s captain speaking. We are expecting to land soon. Sit back and relax. Nothing can go wrong, grr, wrong, wrong, wrong, wrong, …

When it comes to inference in computer science, there is a holy war between the good guys (Bayesian computer scientists) and the bad guys (the notorious frequentist gang). Bayesians try to keep track of the possibility of all different scenarios and how likely they are according to the observations and prior knowledge, and then check to see what the future is going to look like under each scenario (usually in a compact mathematical formulation but with concrete probabilistic interpretation). The frequentist gang, on the other hand, usually take the most likely scenario (or try to accept/reject different scenarios) and use heuristics all around the place, often times without concrete reasoning as to why these heuristics work.

Bayesian inference is, of course, computationally expensive. So the bad guys keep mocking the good guys that their methods are only theoritical and suitable for textbooks rather than real life. Now guess what… They are right! Exact Bayesian inference is not an option in most cases. This means that you need to do approximations. The funny thing is that these approximations usually lead to the exact same heuristics that are used by frequentists. But the good part of going this way is that you know what you are doing and why these methods work. As the great spritual leader Mahdi Milani Fard XIII once said:

Think Bayesian
Act frequentist
And live free

They should write that in gold and put it over the main enterance of all CS departments.

P.S.: This view is of course not exclusively mine. Many CS researchers like to think this way.