Exam 1

Question 1

Data science

Answer 1

An area of investigation which includes AI and it’s components as well as statistical analysis, data analysis

Question 2

Machine learning

Answer 2

Element that allows a computer intelligence to learn

Question 3

Artificial intelligence

Answer 3

Computer implementation of human intelligence

Question 4

Artificial neural network

Answer 4

Computer version of a biological version of this

Question 5

Supervised Learning

Answer 5

Type of ML that find a model based on a dataset where the values (targets) are known, to predict those values

Question 6

Unsupervised learning

Answer 6

Type of ML to find a model based on a dataset to determine natural classifications without guidance from known classifications

Question 7

Classifier

Answer 7

Find a classification

Question 8

Clustering

Answer 8

Find natural classifications for a dataset without guidance

Question 9

Regression

Answer 9

Predict a value based on a fit to trends in the dataset

Question 10

Association

Answer 10

Identify patterns of association between variables or items

Question 11

A learning rule in machine learning is

Answer 11

What ML algorithms uses to learn

Question 12

A decision boundary is

Answer 12

A point, line, plane, or hyper plane separating different classes

Question 13

Gradient descents

Answer 13

Updates the answer in the direction along the negative gradient

Question 14

Steepest descents

Answer 14

Chooses the best learning rate in each step to minimize the number of iterations

Question 15

Choice of optimization

Answer 15

Trade off in number of iterations and the speed of each iteration that produces a reliable result in the shortest time with the smallest resources (memory)

Question 16

The gradient gives

Answer 16

A good direction but not a good distance to find the minimum

Question 17

The problem of diminishing gradients can be handled by

Answer 17

Normalizing the gradient by dividing by its L2 norm

Question 18

L1 Norm of a vector

Answer 18

Sun or average of absolute values of the vector elements

Question 19

L2 Norm of a vector

Answer 19

Square root of sum or average of square of values of the vector elements

Question 20

Linfinity Norm of a vector

Answer 20

The maximum number in the vector

Question 21

L0 Norm of a vector

Answer 21

Number of non zero elements in the vector

Question 22

Feature

Answer 22

Defining characteristics of a given dataset that allow for optimal learning

Question 23

Observation

Answer 23

A sample of the system that may contain several measurements

Question 24

Reason to do feature scaling

Answer 24

Tends to make the search for the minimum more direct

Question 25

K fold cross validation

Answer 25

Divide the data randomly into datasets and choose all combinations to train

Question 26

Overfitting

Answer 26

Fitting noise during training which over estimates how the model will perform on test data

Question 27

Underfitting

Answer 27

Training does not capture the desired prediction

Question 28

Accuracy

Answer 28

Number of correct predictions as a proportion of total observations

Question 29

AUC

Answer 29

Area under the true positive vs False positive rate

Question 30

True positive rate

Answer 30

TP/(TP + FN)

Question 31

False positive rate

Answer 31

FP/(FP + TN)

Question 32

Linear regression

Answer 32

Predicts value from data trends

Question 33

Logistics regression

Answer 33

Generative discriminator classifying using relative probabilities, assumes independent predictors

Question 34

SVM

Answer 34

Margin perceptron with regularizer including linear or nonlinear transform

Question 35

Decision tree

Answer 35

Based on splitting observations on feature value thresholds, a weak learner

Question 36

Random forest

Answer 36

Ensemble learning method using decision trees with randomly chosen datasets

Question 37

K nearest neighbors

Answer 37

Subdivision of space into classes based on majority of members within a distance from a centroid, a lazy learner

Question 38

K means

Answer 38

Unsupervised learning dividing observations into clusters

Question 39

K means with DBSCAN

Answer 39

Clustering based on density of points in a classification

Question 40

Two types of error in ML are

Answer 40

Bias and variance

Question 41

An example of bias would be

Answer 41

An assumption about filtering or collection of data

Question 42

An example of variance would be

Answer 42

Fitting nuances in the data that may be noise

Question 43

A way to reduce variance would be

Answer 43

Filter or smooth the predictors

Question 44

A way to reduce bias would be

Answer 44

Remove filters and use raw data

Question 45

We can generally reduce overfitting by

Answer 45

Increasing the number of independent observations