A simple feedforward neural network
In a simple feedforward neural network, all signals flow in one direction, from input to output. Artificial neurons in the input layer receive signals from the environment and in turn send signals to neurons in the “hidden” layer. Whether any particular neuron sends a signal, or “fires,” depends on the combined strength of signals received from the preceding layer. Output neurons communicate the final result by their firing pattern.
perceptrons
computer science
perceptrons, a type of artificial neural network investigated by Frank Rosenblatt, beginning in 1957, at the Cornell Aeronautical Laboratory at Cornell University in Ithaca, New York. Rosenblatt made major contributions to the emerging field of artificial intelligence (AI), both through experimental investigations of the properties of neural networks (using computer simulations) and through detailed mathematical analysis. Rosenblatt was a charismatic communicator, and there were soon many research groups in the United States studying perceptrons. Rosenblatt...
Discover the world with Britannica
Subscribe to Continue Reading
Free Trial
|
No Ads
|
Exclusive Content
Explore Subscriptions
