Jupyter Notebook 1.2 - Intro to ML

Question 1

What are error functions used for?

Answer 1

They compute the cost of using the current parameters of the model

Question 2

What is a hyperparameter in machine learning?

Answer 2

A hyperparameter is a configuration that is set before the learning process begins and controls the behavior of a machine learning algorithm. Unlike model parameters, which are learned from the data, hyperparameters need to be manually set or tuned.

Examples:
* n_estimators in Random Forest (number of trees)
* learning_rate in gradient descent algorithms
* max_depth in Decision Trees

Tune hyperparameters:
* Grid search
* Random search
* Bayesian optimization

Question 3

What is regularization in machine learning?

Answer 3

Regularization is a technique used to prevent overfitting by adding a penalty term to the loss function. It discourages the model from becoming overly complex by limiting the size of the model’s parameters (weights).