An AI machine learning method that trains a neural network by feeding it predefined sets of inputs. Sometimes used in the pre-training phase but mostly employed when the model is fine-tuned, supervised learning causes the network to learn by example. The neural network is fed pre-labeled inputs and the correct outputs so that the model learns by example. See
AI training and
AI weights and biases.
Unsupervised Learning
In unsupervised learning, there are no labels attached to the input. The AI model detects patterns by recognizing common attributes. Unsupervised learning is less widely used than supervised learning, but both methods may be used to train similar models.
Semi-Supervised Learning
Semi-supervised learning employs both labeled and unlabeled sets of data and is especially useful with huge datasets. Labeling only a small amount of input saves a lot of human time, but it can provide enough samples to let the model identify the larger unlabeled set. See
AI model,
machine learning and
deep learning.