2007-07-19 01:46:42 · 10 answers · asked by ramjan begum 1 in Science & Mathematics ➔ Engineering
For specific problems, Neural Networks (NN) can do very well. They are especially good at tasks like pattern matching, and in this respect start to approach or exceed the human mind. A real example of something a NN has been trained to do better than a human is the game of backgammon. It was good enough that human experts have learned new moves from it.
So NNs can solve very complex problems, but it is still very difficult to apply them to problems that are not well defined. This is general true of all processes used in artificial intelligence. The human being is still king of GENERALIZING, and computers have a long ways to go before humas have to worry about being replaced. It still takes a human to set the problem up right so that a NN can solve it.
But the research in this field is still on-going and in recent years there have been some very solid applications in network routing and language processing (automated translators).
2007-07-19 02:14:14 · answer #1 · answered by Dr. Gene 2 · 0⤊ 0⤋
Research is still going on. Neural network is being applied in vast areas. There will surely be a day when a machine can work like a human brain.
2007-07-19 02:03:24 · answer #2 · answered by ♥Remi♥ 4 · 0⤊ 0⤋
Neural network and fuzzy logic is just the beginning and may be 0.1 % of what human brain can do. (Think, act, sleep etc., etc).
Humans may ultimately develop something which might 'mimic' humans but overtaking humans may take centuries or even more.
Just thinking.... may be yes... Deep Blue, Blue Gene are computers designed to play Chess which beat the World Grand Master.... so it is just the beginning... a long way to go.
2007-07-19 01:57:19 · answer #3 · answered by spansree 2 · 0⤊ 0⤋
I don't know what is fuzzy logic. But I am sure of one thing, to replace human brain networks have to be made in a semi-solid phase and not in solid state. Most of engineering involving hardware is in solid-state. Only when it enters semi-solid state, can it dream of coming near human brain. Another basic criterion is that such a network should work in a differential gaseous milieu. Simply stated, technology has to reach a level wherein it can simulate biological cell to reach the intricacy of human brain. Without that, I don't think we can reach human brain's capacity. However, I am hopeful that technology will soon reach this point. Human creations have some orientational deficit. For e.g. the wheel. We always make circular wheels. So, we rarely think of sliding motion or magnetic motion. Similarly, for us hardware means solid device. Why can't it be semisolid? Just on reading this you might think that semi-solid is unpleasant. That states our prejudice. The day these prejudices die is the day of articial brain.
2007-07-19 21:26:52 · answer #4 · answered by Prakash D 2 · 0⤊ 0⤋
You will always need a human brain to develop and train the neural networks dude .. so east or west human brain is the best !!!!
2007-07-19 01:51:06 · answer #5 · answered by sharad s 2 · 0⤊ 0⤋
Neural community and fuzzy solid judgment is only the beginning and would additionally be 0.a million % of what human innovations can do. (think of, act, sleep and a great form of others., and a great form of others). human beings might probably finally develop some subject which might 'mimic' persons in spite of the shown fact that overtaking persons might probably take centuries or lots greater. only questioning.... would even ensure... Deep Blue, Blue Gene are desktops designed to play Chess which beat the international Grand grasp.... so this is only the beginning... a protracted technique to bypass.
2016-10-22 01:09:36 · answer #6 · answered by zaragosa 4 · 0⤊ 0⤋
A lot of work has been done in this area and it doesn't look too good. If you want to read foundational work on this topic, find a book by Alan Turing who wrote on the topic in the 1940's and for which the Turing competition is held each year. If you want to read something current, pick up anything by Rodney Brooks of MIT. He's also the guy who started the IRobot company.
2007-07-19 01:55:13 · answer #7 · answered by nyquist1995 2 · 0⤊ 0⤋
I know a few people who could benefit from a Neural Network/Brain swap . . .
2007-07-19 07:21:30 · answer #8 · answered by maredudd800 3 · 1⤊ 0⤋
If at all that would have been possible then, this funny question would not have been answered by any of us, but by a robotic supercomputer...got it!
2007-07-19 01:55:29 · answer #9 · answered by presidentofasia 3 · 0⤊ 0⤋
Not if you expect it to do anything useful or unique
2007-07-19 01:52:15 · answer #10 · answered by Gene 7 · 0⤊ 0⤋