Training & Learning
Hyperparameter
A setting you choose before training that controls how the model learns, as opposed to what it learns.
Hyperparameters are the knobs set by the people building a model rather than learned from data, things like the learning rate, the number of layers, or how long to train. They are distinct from parameters, which the model learns on its own. Tuning hyperparameters well can be the difference between a model that works and one that does not, and much of the hands-on craft of training lives here.
Reference: Wikipedia: Hyperparameter
Related terms
Gradient Descent
The optimization method that gradually tweaks a model to lower its error, like walking downhill to the lowest point.
Parameters
The adjustable numbers inside a model that get tuned during training and store what the model has learned.
Epoch
One complete pass of the training process through the entire training dataset.