Neural Networks Explainer
A neural network is a machine-learning model built from layers of small mathematical units often called neurons or nodes. Each connection carries a numerical weight that determines how strongly one signal influences the next. The name is inspired by biology, but an artificial neural network is a computational system rather than a digital copy of the brain.
Input values enter the first layer and are combined with weights and bias terms. An activation function transforms each result before passing it forward through one or more hidden layers. The final layer produces a prediction, such as a class label, a probability, or the next token in a sequence.
During training, the network compares its output with the desired answer and calculates an error. Backpropagation determines how much each parameter contributed to that error, and an optimizer makes small updates. Repeating this process across many examples lets the model learn useful internal representations without a programmer specifying every rule.
Neural networks can model complex nonlinear relationships and power modern vision, speech, and language systems. They can also absorb bias from their data, behave unpredictably on unfamiliar inputs, and require substantial computing resources. Their performance must be measured on representative data, not assumed from a convincing demonstration.
A neural network is a machine-learning model made of connected layers that learn patterns by adjusting numerical weights.
Key terms include node, layer, weight, bias, activation function, forward pass, loss, backpropagation and optimizer.
<a href="https://developers.google.com/machine-learning/crash-course/neural-networks" target="_blank" rel="noopener noreferrer">Google Machine Learning Crash Course: Neural Networks</a> and <a href="https://www.nature.com/articles/nature14539" target="_blank" rel="noopener noreferrer">Nature: Deep Learning</a> provide the primary references for this explainer.
A forward pass moves input through weighted connections and activation functions to produce an output. Multiple hidden layers let the network build progressively more useful representations.
Training uses a loss function to measure error. Backpropagation computes gradients, and an optimizer adjusts the weights so later predictions are more accurate.
Neural networks are flexible enough to learn from images, audio, text, and structured data, but they usually need carefully prepared examples and significant computation.
High test accuracy does not guarantee reliability in a new setting. Teams should check bias, robustness, calibration, privacy, and failure behavior before deployment.
Explore more "Explainers"
Discover additional explainers across politics, science, business, technology, and other fields. Each explainer breaks down a complex idea into clear, everyday language—helping you better understand how major concepts, systems, and debates shape the world around us.