What Is A Neural Network - 1466 Words

What is a Neural Network? Since the birth of the computer era one thing has always been clear: humans and computers are opposites when it comes to strengths in logical thinking. While computers excel in mathematics, solving complex equations faster than a human can start, they fall short in areas like facial recognition and pattern prediction. Computer scientists have made strides in lessening the gap of intelligence between computers and humans, working on several types of programs, referred to as artificial intelligence, to enhance the way computers work and â€Å"think.† One type of AI called an Artificial Neural Network (commonly referred to as either ANN or â€Å"neural net† for short). A neural network is a program that simulates the way a†¦show more content†¦Looking at the diagrams, we can walk through and see how neural networks are modeled after biological neurons. Neurons send and read messages through electrical impulses, much like how computers send and read data through binary. Neurons receive impulses through their dendrites, and send an output impulse through their single axon. Impulses flow between neurons through synapses, connecting one neuron’s dendrites to another s axon. The synapses vary in strength, and either boost or weaken the message being sent to the dendrites, which carry the message towards the cell body to the nucleus. Then, in the cell body, all of the impulses brought in from the dendrites are summed up, and if that sum meets a certain criteria, new impulses is sent along the nuclei axon. In the mathematical model that neural networks use, the process is extremely similar. A variable travels along the axon of a different neuron to the dendrite of the model neuron. The synapses are referred to as weights. The incoming variable is then multiplied by the weight of the path it came through. The values of the weights change depending on how the network is programmed to learn. It is the changing of the weights that is the â€Å"learning† of the network. Then, similar to the biological neuron, all of the inputs are summed up, and if it meets a specific threshold, then the program sends an output. Network Through the Layers In the