AI glossary · Machine learning
Backpropagation
Backpropagation is how a neural network learns from mistakes: it works backwards from the error to find how much each connection caused it, then adjusts them.
Also called backprop
In more detail
After the network makes a prediction and the loss function measures the error, backpropagation traces that error back through each layer. This tells gradient descent which way to move every weight.
Popularised in the 1980s, it is still the foundation of how deep learning models are trained.
Related terms
Gradient descentGradient descent is the method most AI models use to learn: it repeatedly nudges the model’s settings in the direction that reduces its errors.Neural networkA neural network is an AI model made of layers of connected units, loosely inspired by the brain, that learns by adjusting the strength of its connections.Loss functionA loss function measures how wrong an AI model’s predictions are during training. Training tries to make this number as small as possible.Deep learningDeep learning is a type of machine learning that uses neural networks with many layers to learn complex patterns in text, images and sound.
Browse related tools
AI tools listed on BroUseAI in these categories.
