Artifical Intelligence Essays - Computational Neuroscience

Artifical Intelligence Essays - Computational Neuroscience Artifical Intelligence Artificial Intelligence Artificial intelligence is a highly debatable topic. You either believe that it may be achieved or think it can't, and the middle is a little shady. Artificial Intelligence is the study to create a machine that can act like a human brain, including emotions, and consciousness. This speech will cover the subject of if it can ever be achieved and at what level. This would be a giant technological step. If it is ever achieved, everyday activities such as vacuuming, or laundry, would become automated. The leader in the field of AI is actually not a business, but MIT's Artificial Intelligence Laboratory. It does more ground breaking research in the all fields of AI, including robotics and neural networks, than any other facility in the world. First I will describe the different techniques and fields that fall under AI. The first technique that is used to create intelligent systems is the top-down approach. That is when the computer is given a written set of directions such as IF that is red, THEN go forward and it will go from there according to the program. This technique was widely used back in the 1950's when AI was still a mysterious subject, but it has major flaws. To write complex programs that take in all their surroundings would take hundreds of thousands of lines of code, which no one has the time to write. The newest technique that is being used is bottom-up. That is the arrangement of data into a complex system where all information is connected, like a brain, which is why it is also referred to as a neural network. It is extremely complex and confusing at some points, so I won't describe it here, but you don't need to understand it to answer the question at hand. There are also many different fields of study that fall under AI. The first field that I'll talk about is robotics. This field has been prospering in the '90's. It is not so much the engineering for the robots that is causing the standstill in functionality of these machines, but the computers that control what they do, which is the problem with all forms of AI right now. There is also database processing, which works with the way that a computer sorts and retrieves information. And intelligent software agents that can tell you when to buy and sell stock among many other things. But, what this speech mainly focuses on is the theory behind all this, which has its own field. There are four categories of viewpoints that one may have on this topic. A. All thinking is computation; in particular, feelings of conscious awareness are evoked merely by the carrying out of appropriate computations. B. Awareness is a feature of the brain's physical action; and whereas any physical action can be simulated computationally, computational simulation cannot by itself evoke awareness. C. Appropriate physical action of the brain evokes awareness, but that physical action cannot even be properly simulated computationally. D. Awareness cannot be explained by physical, computational, or any other scientific terms. Penrose 12 All that says is that it all depends if the brain functions computationally or physically (chemical reactions) and whether or not you believe that either of those can be simulated by a machine or biological creation of some kind. We will be dealing with viewpoint A and D. If you believe in viewpoint A than you believe that computers can become aware of their surroundings. The word aware means that it can take in its whole surroundings and make judgments and actions for itself, which a giant leap from where AI is at now. It really depends on how you view how the brain functions. Research is being done by neurobiologists to unlock the secret of how the brain works. It could function by computation or chemical reactions, no one is really sure. If you think that it is all done computationally, and that humans can simulate it in a machine of some sort than you believe in A. A is an implication of a highly operational attitude to science, where, also, the physical world is taken to operate entirely computational. Penrose 13. There has been no evidence that supports or denies this viewpoint because the secret of how the brain works has not been solved, yet. The viewpoint D, on the other hand totally rejects the whole idea that computers, or any machine can achieve any form of intelligence that may be mistaken for human action. The main reason for believers in this category is from the work of a German mathematician named Kurt Goedel. He was