BANA 6660 Quiz 4 - Bootstrap, Random Forests and Boosting

0.0(0)
Studied by 0 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/7

encourage image

There's no tags or description

Looks like no tags are added yet.

8 Terms

1
New cards

How does a random forest work?

Combines trees built with bootstrap samples and random predictor subsets.

2
New cards

How does boosting differ from random forest?

Builds successive trees that focus on earlier errors.

3
New cards

What does bootstrap sampling with replacement mean?

The same observation can be selected multiple times.

4
New cards

What does 63.2% mean in bootstrap sampling?

About 63.2% of original observations appear at least once, on average.

5
New cards

What does 36.8% mean in bootstrap sampling?

About 36.8% of original observations are left out, on average.

6
New cards

What is out-of-bag (OOB) validation?

Evaluating trees with observations omitted from their bootstrap sample.

7
New cards

What problem do random forests mainly reduce?

Variance, by combining predictions from many trees.

8
New cards

What problem does boosting mainly reduce?

Bias, by sequentially targeting earlier errors.