Cue topics TT2 CSC311

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

1/23

flashcard set

Earn XP

Description and Tags

contains unit from 5-9 with highlevel ideas

24 Terms

1
New cards
term image

2 layer MLP for XOR -6

2
New cards

1. Why is a deep linear network no more expressive than a linear model?

By repeated substitution, any number of linear layers collapses to one.

- deep network whose hidden activations are all linear has the same expressiveness as logistic/ linear regression. - 6

3
New cards
term image

non vectorized back prop -6

4
New cards
term image

vectorized back prop - 6

5
New cards
term image

bias variance decomposition -7

6
New cards
term image

bagging - 7

7
New cards
term image

Maxium likelihiid estimation - univariate - 7

8
New cards
term image

Maxium A posteriori ; univariate 7

9
New cards
term image

Binary Bag of Words - 8

10
New cards
term image

Naive Bayes assumption -8

11
New cards
term image

generative classifier - 8

12
New cards
term image

Compute MLE Naive bayes - 8

13
New cards
term image

Derive MLE naive bayes classifer -8

14
New cards
term image

Inference in a Naive Bayes Classifier-8

15
New cards
term image

Computing MAP of Parameters of Naive Bayes Classifief=r - 8

16
New cards
term image

GDA with One Feature - 9

17
New cards
term image

Computing MLE of Parameters of a Multivariate Gaus-

sian - 9

18
New cards
term image

Computing MLE of Parameters of GDA-9

19
New cards
term image

GDA with Two Features - 9

20
New cards
term image

Drawing the Gaussian contours - 9

21
New cards
term image

Linear seperability -5

22
New cards
term image

Multilayer Perceptrons - 5

23
New cards
term image

Inference 3 layer MLP - 5

24
New cards
term image

Decision Boundary of Linear Classification Model - 5