SIGNAL//DESK
AI/MLsrc: curated AI glossary

Neural network

A neural network is a computer system inspired by the human brain that learns to recognize patterns by looking at many examples, gradually adjusting its internal settings until it gets the right answer.

A computational model composed of interconnected layers of nodes that process data; it learns to map inputs to outputs by iteratively adjusting the weights of these connections based on feedback from its own performance.

A layered system of weighted connections ('neurons') that transforms inputs to outputs; the weights are learned from data by minimising error via gradient descent and backpropagation.

evolution

  1. 1943 · history
    McCulloch-Pitts Neuron

    The first mathematical model of a biological neuron was proposed to simulate logical functions.

  2. 1958 · history
    The Perceptron

    Frank Rosenblatt created the first artificial neural network capable of learning through trial and error.

  3. 1986 · history
    Backpropagation Popularization

    Rumelhart, Hinton, and Williams demonstrated that backpropagation could efficiently train multi-layer networks.

  4. 2012 · history
    AlexNet Breakthrough

    A deep convolutional neural network won the ImageNet competition by a massive margin, triggering the modern deep learning era.

  5. 2017 · history
    Transformer Architecture

    The introduction of the Transformer model shifted the field toward attention mechanisms, enabling large-scale language modeling.


← all terms