Hyperparameter
A configurable value set before training that controls how a machine learning model learns, such as learning rate or batch size.
What is a Hyperparameter?
Hyperparameters are defined before or during the model training process and influence how the model learns. Examples include the learning rate, batch size, number of training epochs, number of hidden layers, and regularization strength. Unlike model parameters such as weights and biases, hyperparameters are typically selected by developers or determined through tuning techniques.
Why is a Hyperparameter Important?
The choice of hyperparameters can significantly affect a model’s accuracy, training speed, generalization, and computational requirements. Poor settings may lead to problems such as overfitting, underfitting, or inefficient training. Hyperparameter tuning helps identify configurations that produce better model performance.
Common use cases
Hyperparameters are commonly used in neural network training, classification, regression, deep learning, model optimization, and machine learning experimentation.