Chapter 1

Question 1

What is Supervised Learning?

Answer 1

When training data is fed with labels that indicates the solutions (contains y in train)

Question 2

Name some Supervised Learning Algorithms.

Answer 2

KNN
Linear Regression
Logistic Regression
Support Vector Machines
Décision trees and Random Forests
Neural Networks (sometimes)

Question 3

What is Unsupervised Learning?

Answer 3

The training data is unlabeled (no y), the class tries to learn without a teacher

Question 4

Name some Unsupervised Learning Algorithms

Answer 4

Through clustering:
KMeans
DBSCAN
Hierarchical Cluster Analysis
Through Anomaly detection:
One class SVM
Isolation Forest
Visualization and Dimensionality Reduction:
Principal Component Analysis
Kernel PCA
Locally Linear Embedding
T-distributed Stochastic Neighbor Embedding
Association rule learning:
Apriori
Eclat

Question 5

What is classification?

Answer 5

Examples are with their class in order to classify new emails

Question 6

What is Regression?

Answer 6

Predicting a target numeric value given a set of features called predictors. Training a model requires both predictors and labels.

Question 7

What is a clustering algorithm?

Answer 7

An algorithm to detect similarities of data points based on feature combos.

Question 8

What is hierarchical clustering?

Answer 8

Subdivision of a clustering algorithm into smaller groups

Question 9

What is a visualization algorithm?

Answer 9

An algorithm that outputs a 2d or 3d representation of data that can be plotted easily.

Question 10

What is Dimensionality Reduction?

Answer 10

Simplifying data without losing too much data,trying to merge many correlated features into one.

Question 11

What is feature extraction?

Answer 11

Merging multiple features into one

Question 12

Should you reduce the dimensions of data before feeding it into a Supervised ML algorithm?

Answer 12

Yes, it will likely perform better and quicker while reducing strain on storage and processing

Question 13

What is Anomaly detection?

Answer 13

A model that takes in normal data and removes or flags any with a very different result, usually used to remove outliers.

Question 14

What is novelty detection?

Answer 14

The same as Anomaly detection but they only see normal data , no outliers

Question 15

What is association rule learning?

Answer 15

Looking through large amounts of data and discover new relations between attributes only possible with enough data.

Question 16

What is Machine Learning?

Answer 16

The science and art of programming computers so they can learn from data

Question 17

What is Machine Learning?

Answer 17

The science and art of programming computers so they can learn from data

Question 18

What is semisupervised learning?

Answer 18

Algorithms that use a lot of unlabeled data to group, then a little labeled data to classify the whole collection

Question 19

Describe a Deep Belief Network (DBN)

Answer 19

Unsupervised components called Boltzmann Machines (RBMs) stacked on top of each other.thenwjole system is trained unsupervised and then fine tuned using supervised techniques

Question 20

What is an Agent in Reinforcement learning?

Answer 20

The learning system that can observe the environment, select and perform actions, and get rewards or penalties. It must then learn a policy l

Question 21

What is a policy in Reinforcement learning?

Answer 21

A policy defines what action the agent should choose when it is in a given situation.

Question 22

What is batch learning?

Answer 22

Batch learning is when a system cannot learn incrementally and must learn on all available data.

Question 23

Describe the process of offline Learning?

Answer 23

System is first trained on batch learning, offline and then it is launched into production without learning anymore

Question 24

For predicting stock prices, which would be better and why: offline learning or online learning?

Answer 24

Online learning, as it is done incrementally, stock data can be trained in small amounts to react quickly to.the change in data

Question 25

What is out of core learning?

Answer 25

Using online learning algorithms to train systems on huge datasets that cannot fit on one machines main memory.

Question 26

What is a learning rate?

Answer 26

A learning rate adjusts how fast a system adapts to new data. The lower the threshold, the more resilient it is to change.

Question 27

What is a utility function?

Answer 27

A measure of how correct a model is

Question 28

What is a cost function?

Answer 28

A measurement of how incorrect a model is

Question 29

Train and predict using a linear regression model in scikit-learn

Answer 29

Must have:
Import sklearn.linear_model

X = data.drop(columns = “target”)
Y = data[‘target’]

Model = sklearn.linear_model.LinearRegression()

Model.fit(X,y)

X_new = [[new data matching x]]
Model.predict(X_new)

Question 30

What is inference?

Answer 30

Predicting based on an algorithm

Question 31

What is the issue with nonrepresentitve training data?

Answer 31

The data will only reflect a population that is unlikely to create an accurate generalization

Question 32

What is sampling bias?

Answer 32

When the method of sampling is flawed and biases the data

Question 33

What are some options when dealing with significant missing values?

Answer 33

Ignore the feature, ignore the missing values, impute, or train models with and without it.

Question 34

What is feature extraction?

Answer 34

Feature extraction is creating the most relevant features from the total features

Question 35

What is overfitting?

Answer 35

Overfitting is when a model becomes too biased to a training set

Question 36

How to solve overfitting?

Answer 36

Simplifying the model, gather more training data, and reduce noise in the training data

Question 37

What is regularization?

Answer 37

Constraining a model to make it simpler to avoid overfitting

Question 38

What is under fitting?

Answer 38

Opposite of overfitting, the model is too simple to learn the data structure

Question 39

How to solve under fitting?

Answer 39

Select a more powerful model, feeding better features, reducing the regularization

Question 40

What is holdout validation?

Answer 40

When training a few models: label some of the training set as a validation set, train multiple models with hyperparameter tuning, test on the validation set, and then train the best on both the validation and training set to be used for the test set.

Question 41

What is cross-validation?

Answer 41

The model is evaluated on several small validation sets with the average being representative of it’s score

Question 42

What is the No Free Lunch theorem?

Answer 42

If you make no assumptions of the data,you should have no preference for model