Concept
Neural network
A computational model built from layers of simple connected units whose weighted connections are learned from data.
Updated Nov 2, 2026 · 1 min read · 6 connections
- First model
- McCulloch and Pitts, 1943
- Perceptron
- Rosenblatt, 1958
- Training
- Backpropagation
Origins
Artificial neural networks were inspired by the biological neuron. Warren McCulloch and Walter Pitts described a mathematical model of a neuron in 1943, and Frank Rosenblatt built the perceptron in 1958.
Structure and training
A network is a weighted, directed structure of the kind studied in graph theory. Training adjusts its weights to minimize a loss function, often cross-entropy, a quantity drawn from information theory and probability theory.
Whether such systems could ever have consciousness remains an open question.