What is a hyperparameter in machine learning?

A hyperparameter in machine learning is indeed a configuration setting that is external to the model. These settings are not learned from the training data during the model training phase; instead, they are specified before the learning process begins and can significantly impact the performance of the model. Hyperparameters include factors like the learning rate, the number of trees in a random forest, or the number of layers and units in a neural network.

These settings influence how the model learns from the training data and how it generalizes to unseen data. Since hyperparameters are set prior to training, selecting the right values often requires experimentation and tuning, which can be done through techniques like grid search or random search to find optimal configurations.