Research Paper on Neural Network
The study of the human brain has been researched for thousands of years. As technology progressed it seemed only natural to try and replicate the thought process. A Neurophysiologist, Warren McCulloch, and a mathematician, Walter Pitts researched how neurons work in 1943. They modeled a basic neuron by electrical circuits. This was considered the first step toward the simulation of the neural network. By definition, neural networks are relatively crude electronic models based on the neural structure of the brain.
____________________________________________________________________________________________________________________
The questions that are being developed and answered are:
1. How does the neurons/neural network work?
2. How are they being used? And more importantly, how efficient is the neural network?
Theories concerning replicating the human brain formalized again in the 1950's, when computers started becoming advanced. An engineer, Nathaniel Rochester from IBM research laboratories began the first effort to simulate a true neural network. Of course, with most first attempts, they fail. And that was the case with Rochester. But, as he continued in his research they became successful.
In 1959, Bernard Widrow and Marcian Hoff developed models called ADALINE and MADALINE.
The names are acronyms for their use of Mulitple Adaptive LINear Elements. It’s an adaptive filter that removes the echoes from phone lines. MADALINE was considered the first neural network to actually be the solution for a true problem. The neural network is still currently being used in some commercial businesses. Through the years, the success turned to fear and drew many concerns of the accountability of the machines thinking and resolving problems. Due to the excessive hype of uncertainty, much of the funding came to a halt.
In 1982, John Hopfield proposed to create valuable devices, by using neural networks. Since then several engineers and scientists have come from around the world to Neural Network conferences to discuss ideas, and engaged in targeting specific uses for the network. As progression took place engineers realized that the key to the neural networks lies in hardware development. Scientists came to the conclusion that in order to fully develop the networks the prototypes had to be removed from the labs for testing. Which required specialized chips. Companies began working on three types of neuro chips- digital, analog, and optical. Currently, the optical chip seems to be the most promising for the future.
The questions that are being developed and answered are:
1. How does the neurons/neural network work?
2. How are they being used? And more importantly, how efficient is the neural network?
Theories concerning replicating the human brain formalized again in the 1950's, when computers started becoming advanced. An engineer, Nathaniel Rochester from IBM research laboratories began the first effort to simulate a true neural network. Of course, with most first attempts, they fail. And that was the case with Rochester. But, as he continued in his research they became successful.
In 1959, Bernard Widrow and Marcian Hoff developed models called ADALINE and MADALINE.
The names are acronyms for their use of Mulitple Adaptive LINear Elements. It’s an adaptive filter that removes the echoes from phone lines. MADALINE was considered the first neural network to actually be the solution for a true problem. The neural network is still currently being used in some commercial businesses. Through the years, the success turned to fear and drew many concerns of the accountability of the machines thinking and resolving problems. Due to the excessive hype of uncertainty, much of the funding came to a halt.
In 1982, John Hopfield proposed to create valuable devices, by using neural networks. Since then several engineers and scientists have come from around the world to Neural Network conferences to discuss ideas, and engaged in targeting specific uses for the network. As progression took place engineers realized that the key to the neural networks lies in hardware development. Scientists came to the conclusion that in order to fully develop the networks the prototypes had to be removed from the labs for testing. Which required specialized chips. Companies began working on three types of neuro chips- digital, analog, and optical. Currently, the optical chip seems to be the most promising for the future.
___________________________________________________________
Warning!!! All free online research papers, research paper samples and example research papers on Neural Network topics are plagiarized and cannot be fully used in your high school, college or university education.
If you need a custom research paper, research proposal, essay, dissertation, thesis paper or term paper on your topic, EffectivePapers.com will write your research papers from scratch. Starting at $12/page you can order custom written papers online. We work with experienced PhD. and Master's freelance writers to help you with writing any academic papers in any subject! High quality and 100% non-plagiarized papers guaranteed!
___________________________________________________________Warning!!! All free online research papers, research paper samples and example research papers on Neural Network topics are plagiarized and cannot be fully used in your high school, college or university education.
If you need a custom research paper, research proposal, essay, dissertation, thesis paper or term paper on your topic, EffectivePapers.com will write your research papers from scratch. Starting at $12/page you can order custom written papers online. We work with experienced PhD. and Master's freelance writers to help you with writing any academic papers in any subject! High quality and 100% non-plagiarized papers guaranteed!
