Introduction to machine learning · Question 45 of 60
Which examples belong to model hyperparameters and which to algorithm hyperparameters?
3 answers are correct. Hard
Show answer
Answer
- A Depth of a decision tree: model hyperparameter
- C Number of layers of a neural network: model hyperparameter
- E Number of training iterations: algorithm hyperparameter
Explanation
Model hyperparameters determine the structure, for example how many layers a network has or how deep a decision tree may grow. Algorithm hyperparameters steer learning, for example the number of passes. During tuning, both are adjusted depending on the evaluation, and then the model is retrained.