The Difference Between Memorizing and Learning

Assalamualaikum wbt,

Today I learnt new thing.
This is actually from my supervisor.

When we were discussing about how computer could learn,
He gave a simple example of a teacher and a pupil.

Say a techer tries to teach his pupil arithmatic of adding.
So he told his pupil that:
1+1 = 2
1+2 = 3
1+3 = 4
...

Alrite, done with the session.
Now if he wanted to know if the pupil memorized it,
He could ask:
1+1 = ?
1+2 = ?
1+3 = ?

Obviously, the pupil could answer these equations if he had memorized what the teacher was teaching.

But if he wanted to test if the pupil learned 'adding' or not,
then he sould ask a different set of questions, say:
5+6 = ?
3+7 = ?
1+8 = ?
...

If the pupil can answer the test correctly, then we know the pupil had actually Learned the arithemetic of adding.
If he could answer 1+1, 1+2, 1+3, but could not answer a new set of questions, it means he had not learned the arithmetic but simply memorized what the teacher had tought him.

So...this is actually a simple way to describe the difference between memorizing and learning.

Now we memorized this kalimah La ilaaha illallah, Muhammad ur-Rasulullah.
We say this kalimah everyday, time after time.
But,
have we learn it?

ps: kita 'train' dalam solat, nak 'test' solat..tgk sikap kita kat luar solat.

Sekian...
-Nazri-

Cermin

Assalamualaikum,

Kekadang kita lupa diri.

Rambut tak kemas. Kita selalu tak sedar.
Pipi ada kotor, kita selalu tak sedar.
Sedar bila tengok reflection diri depan cermin.

Macam tu la jugak lebih kurang peribadi kita nih.
Sifat-sifat yang kurang baik.
Selalunya kita tak sedar.
Dan cermin nya?
Itulah sedara-sedara muslim kita yang lain.
Mereka lah cermin kita.



.

Mata lagi..

Saya tak tahu mengapa,
hari ni saya macam bersemangat untuk mengepost di blog ni.

Mana tak nya, arini aku cume menunggu program komputer ni habis mengira.
Kami run Genetic Programming untuk improve Sammon Mapping yang terkenal dengan multidimensional mapping tekniknya.

This GP basically tries to find the best function to describe the mapping from multidimensional data to a plane (2-Dimensions).
Macam mana nak tau dia best ke tak?
So, dlm kes ni Sammon punye equation tetap digunakan iaitu menggunakan satu algorithm untuk menilai sejauh mana pemetaan baru dapat mengekalkan jarak Gaussian antara titik.
Algorithm nye malas lah aku nak salin kat sini, sbb ade banyak simbol math..payah nak taip.

It somehow, works! Alhamdulillah, but I cant actually draw the conclusion that this GP teknik is better than the classical Sammon Mapping. Because...the GP runs with many parameters. So, it really depends on your luck, on how you could alter these parameters.

Anyhow, kat sini manusia boleh berfikir, sejauh mana teori Darwin ttg evolusi benar atau salah.
Macam mana lah dia boleh kata kita daripada monyet ye? Aku buat GP ni, nak hasilkan satu function untuk map benda yang mudah pon macam mustahil! Inikan pula nak ubah DNA beruk kepada DNA manusia!

Maaf cakap, tapi sebenarnya memang akal kita ni terhad. Fikir lah dalam sedalam lautan pun, lautan tetap takkan mampu kau duga. Fikirlah tinggi setinggi langit pun, langit takkan mampu kau teroka.

Hmm..back to my project, actually there's a reason why we reduce the dimension of the datasets. The reason which is the results from human defect. They cant see more than 3 dimensions. Hoho! Kita memang buta.

Sedangkan Allah mengetahui apa-apa yang ghaib dan yang nyata.