Gunakan MimeTex/LaTex untuk menulis simbol dan persamaan matematika.

Welcome to Forum Sains Indonesia. Please login or sign up.

Maret 28, 2024, 06:35:15 PM

Login with username, password and session length

Topik Baru

Artikel Sains

Anggota
Stats
  • Total Tulisan: 139,653
  • Total Topik: 10,405
  • Online today: 112
  • Online ever: 1,582
  • (Desember 22, 2022, 06:39:12 AM)
Pengguna Online
Users: 0
Guests: 109
Total: 109

Aku Cinta ForSa

ForSa on FB ForSa on Twitter

Artificial Neural Network

Dimulai oleh reborn, November 30, 2006, 09:00:26 AM

« sebelumnya - berikutnya »

0 Anggota dan 2 Pengunjung sedang melihat topik ini.

reborn

Thread ini mudah-mudahin bisa dipakai buat bahas tentang Neural Network. Ini sedikit pendahuluan dulu tentang topik ini. Terjemahan dan pembahasan lebih lanjutnya menyusul  ;D

What Is Artificial Neural Network?

An Artificial Neural Network (ANN) is an information processing paradigm that is inspired by the way biological nervous systems, such as the brain, process information.

The key element of this paradigm is the novel structure of the information processing system. It is composed of a large number of highly interconnected processing elements (neurones) working in unison to solve specific problems. ANNs, like people, learn by example. An ANN is configured for a specific application, such as pattern recognition or data classification, through a learning process. Learning in biological systems involves adjustments to the synaptic connections that exist between the neurones. This is true of ANNs as well.


Why use neural networks?

Neural networks, with their remarkable ability to derive meaning from complicated or imprecise data, can be used to extract patterns and detect trends that are too complex to be noticed by either humans or other computer techniques. A trained neural network can be thought of as an "expert" in the category of information it has been given to analyse. This expert can then be used to provide projections given new situations of interest and answer "what if" questions.

Other advantages include:

1. Adaptive learning: An ability to learn how to do tasks based on the data given for training or initial experience.
2. Self-Organisation: An ANN can create its own organisation or representation of the information it receives during learning time.
3. Real Time Operation: ANN computations may be carried out in parallel, and special hardware devices are being designed and manufactured which take advantage of this capability.
4. Fault Tolerance via Redundant Information Coding: Partial destruction of a network leads to the corresponding degradation of performance. However, some network capabilities may be retained even with major network damage.


Neural networks versus conventional computers

Neural networks take a different approach to problem solving than that of conventional computers. Conventional computers use an algorithmic approach i.e. the computer follows a set of instructions in order to solve a problem. Unless the specific steps that the computer needs to follow are known the computer cannot solve the problem. That restricts the problem solving capability of conventional computers to problems that we already understand and know how to solve. But computers would be so much more useful if they could do things that we don't exactly know how to do.

Neural networks process information in a similar way the human brain does. The network is composed of a large number of highly interconnected processing elements(neurones) working in parallel to solve a specific problem. Neural networks learn by example. They cannot be programmed to perform a specific task. The examples must be selected carefully otherwise useful time is wasted or even worse the network might be functioning incorrectly. The disadvantage is that because the network finds out how to solve the problem by itself, its operation can be unpredictable.

On the other hand, conventional computers use a cognitive approach to problem solving; the way the problem is to solved must be known and stated in small unambiguous instructions. These instructions are then converted to a high level language program and then into machine code that the computer can understand. These machines are totally predictable; if anything goes wrong is due to a software or hardware fault.

Neural networks and conventional algorithmic computers are not in competition but complement each other. There are tasks are more suited to an algorithmic approach like arithmetic operations and tasks that are more suited to neural networks. Even more, a large number of tasks, require systems that use a combination of the two approaches (normally a conventional computer is used to supervise the neural network) in order to perform at maximum efficiency.

st


reborn

Kutip dari: st pada Desember 17, 2006, 01:45:44 AM
Mantab min, jago euy... :)

hehe... lanjutin dong mod. itu kan baru Intro  :P pembahasan detil Neural Network nya sih ga ngerti  :-[

advisor

Ditunggu lanjutannya... pengen tau juga nehh. Kalo sampe manusia bisa bikin mesin yang bisa belajar sendiri wuihhh.... gmn tuh... ikut2an atheis kali yahh  ;D

punkoholic

glek.... awak cuma bisa menelan ludah ...

:o

jesuisnoel

Stau gw...adaptive learning di ANN blum ada implementasinya. Sejauh ini result yang baru ada ya, Supervised Learning (dimana termasuk didalamnya poin 2-4).

Supervised Learning adalah mostly used learning type, dimana kita harus training system dengan berbagai macam input trus dibandingkan dengan expected output. Kalo ada eror/penyimpangan, kita harus adjust sampai sejauh mana sistem mentolerir adanya penyimpangan itu. Contoh: pattern recognition (kaya di fingerprint/handwritting recognition). Di HWR applikasi harus diajarin berbagai macam huruf A (dari 'A' orang yg tulisannya rapi, sampai yg kaya dokter :P)...di fingerprint lebih repot lagi..sistem mesti bisa 'menyaring' gambar sidik jari yang diinput (either ada org iseng yg nekan jarinya kuat2 waktu input or ada orang bersidik jari halus bgt) dgn ngilangin atau ngurangin noise...baru image yang udah bersih di proses entah dengan neural network, fuzzy lgic algorithm, ato apalah...untuk bisa mendeteksi dan mentolerir minutiae2 yg diterima..

Kenapa unsupervised/adaptive learning masih blm bisa diterapkan?
Mungkin karena terlalu banyak if-else yang bisa terjadi dan belum tentu bisa di handle kali ya...ngga ada patokan pasti gimana sistemhaus bersikap pada saat dia menghadapi lingkungan baru. Analoginya: kita aja panik kl bangun2 rumah udah banjir 26 cm...ngga tau mesti ngapain. Apa lagi robot...hehe  ;D

Mudah2an research tentang ini terus berlanjut.... Ato ada yg mau/sedang jadiin research topic buat PhD mungkin?  ;)

DiaB10

mantap min...lanjutin lagi donk...:D

idiotique_hebb

Kutip dari: jesuisnoel pada Februari 08, 2007, 11:23:41 PM

Kenapa unsupervised/adaptive learning masih blm bisa diterapkan?
Mungkin karena terlalu banyak if-else yang bisa terjadi dan belum tentu bisa di handle kali ya...ngga ada patokan pasti gimana sistemhaus bersikap pada saat dia menghadapi lingkungan baru.

Perbedaan penggunaan learning type antara supervised learning dan unsupervised learning yaitu sisi permasalahannya.
Supervised learning lebih banyak digunakan untuk linear problem contohnya adalah algoritma2 yang menggunakan MLP (Multi Layer Perceptron) seperti Backpropagation, Quickprop, Hebbian, Adaline, Linear Vector Quantization (LVQ) dll.
Sedangkan Unsupervised Learning banyak menggunakan jaringan2 competitive seperti Kohonen, ART1, ART2, Fuzzy ART, Hopfield dll.

Algoritma2 ini jauh lebih elegan daripada metode if-else, jaringan bisa menggunakan metode fuzzy untuk pengendalian pembelajaran maupun output, sehingga bisa menghasilkan output berdasarkan fakta2 yang tidak lengkap.
Lagi belajar Objective C / GNUStep / Cocoa API
Kroper for Mac : [pranala luar disembunyikan, sila masuk atau daftar.]

idiotique_hebb

Tambah lagi :
Permasalahan sistem harus bersikap pada saat dia menghadapi lingkungan baru, (disebut juga generalisasi). Sebuah sistem berbasis ANN dituntut untuk mampu mengenali pola dari learning set dan mampu menghasilkan output yang benar pada saat mencoba mengenali pola lain tanpa harus ada target. Nah, proses ini lah yang disebut generalisasi, sistem "dipaksa" untuk memproses data yang benar2 baru. Berdasarkan hasil uji coba komputasi yang telah dilakukan, projection generalizing neural networks tidak selalu memberikan kapabilitas generalisasi yang lebih baik. projection generalizing neural networks memberikan kapabilitas generalisasi yang lebih baik ketika jumlah data pembelajaran cukup kecil atau variansi noise dari data pembelajaran cukup besar.
Lagi belajar Objective C / GNUStep / Cocoa API
Kroper for Mac : [pranala luar disembunyikan, sila masuk atau daftar.]