L9a: Ensemble Methods

Question 1

What is a decision tree?

Answer 1

Question 2

What is a decision tree?

Answer 2

Question 3

How do you compute the weighted gini at each split?

Answer 3

Question 4

What can Decision Trees be used for?

What is the regression problem?

What is the main hyperparameter of interest?

Answer 4

Question 5

What are the Advantages of using decision trees?

Answer 5

Question 6

What are some disadvantages of using decision trees?

Answer 6

• Not that robust

Question 7

What are the two emsemble methods we wil be looking at?

Answer 7

Question 8

What is the Random Forest emsemble method?

Answer 8

Question 9

What is Bootstrap sampling?

Answer 9

Question 10

What does the Random Forest Classifier look like?

Answer 10

• alot more robust

Question 11

What are the key hyperparameters that need to be tuned in the Random Forest method?

What are some additional hyperparameters?

Answer 11

Question 12

What is the Gradient Boosting ensemble method?

What form does the model take?

Answer 12

Question 13

What are the three importants that have been made to Gradient Boosting?

Answer 13

1. Stochastic Gradient Boosting
2. XGBoost (eXtreme Gradient Boosting)
3. Histogram-Based Gradient Boosting (HGBT)

Question 14

What is Stochastic Gradient Boosting?

Answer 14

Question 15

What is XGBoost?

Answer 15

Question 16

What is Histogram-based Gradient Boosting (HGBT)

Answer 16