Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Introduction to Machine Learning > AWS Machine Learning Foundations Course - Lesson 2 > Flashcards

AWS Machine Learning Foundations Course - Lesson 2 Flashcards

1
Q

What does log loss seek to do

A

Calculate how uncertain your model is about the predictions it is generating - how likely a model thinks the predictions being generated are to be correct

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

how would you define Hyperparameter

A

Settings on the model which are not changed during training, but can affect how quickly or how reliably the model trains

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What does a FFNN do

A

structures neurons in a series of layers, with each neuron in a layer containing weights to all neurons in the previous layer

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

How would you define discrete

A

refers to the outcome taking on only a finite number of values like days of the week

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

How would you define data vectorization

A

Process that converts non-numeric data into a numerical format, so that it can be used by a machine learning model

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

How would you define clustering

A

Helps to determine if there are any naturally occurring groups in the data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

How would you define accuracy

A

The fraction of predictions a model gets right

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What does CNN represent

A

Nested filters over grid-organized data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What are CNNs most used for

A

Processing images

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

How would you define a continuous label

A

Does not have a discrete set of possible values, the label, in theory, could be anything

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

How would you define a categorical label

A

Has a discrete set of possible values

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What does bag of words do

A

Counts how many times a word appears in a document (corpus) and then transforms that information into a dataset

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

How would you define bag of words

A

technique used to extract features from the text

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

How would you define loss function

A

measurement of how close the model is to its goal

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What are some aspects that can ultimately be the largest factor that affects how well you can expect your model to perform

A
  1. Outliers
  2. Missing or incomplete values
  3. Data that needs to be transformed or preprocessed so its in the correct format to be used by the model
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is the fundamental question to ask for data collection

A

Does the data I have collected match the machine learning task and problem I have defined

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What are the four aspects of working with data

A
  1. Data collection
  2. Data inspection
  3. Summary Statistics
  4. Data Visualization
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

If your data is categorical, what task would you be working with

A

Classification

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

If your data is numerical, what task would you be working with

A

Continuous

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

What is a deep learning model composed of

A

Collections of neurons connected together by weights

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What is a tree based model

A

They learn to categorize or regress by building an extremely large structure of nested if/else blacks, splitting the world into different regions at each block

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

What does training determine in tree based models

A

Where splits happen and what value is assigned at each leaf region

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

What are linear models good for

A

Giving a baseline against which to compare more complex models

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

What is a linear model

A

The relationship between a set of input numbers and a set of output numbers through a linear function

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

What is the end-to-end training process

A
  1. Feed the training data into the model
  2. Compute the loss function on the results
  3. Update the model parameters in a direction that reduce loss
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

What type of algorithm would you use to segment your customers into multiple groups

A

If you don’t know how to define the groups, then you can use a cluttering algorithms (unsupervised learning) to segment your clusters of similar customers

If you know what groups you would like to have, then you can feed many examples of each group to a classification algorithm (supervised learning) and it will classify all your customers into these groups

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

What type of machine learning algorithm would you use to allow a robot to walk in various unknown terrains?

A

reinforcement learning

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

Do you want the RMS (root mean square) to be high or low

A

Low

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
29
Q

How would you define root mean square (RMS)

A

Roughly the average error across the test dataset, in general as the model improves, the better the RMS result will be

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
30
Q

Would you frame the problem of spam detection as a supervised leaning problem or an unsupervised learning problem

A

Spam detection is a typical supervised learning problem: the algorithm is fed many emails along with their labels (spam not spam)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
31
Q

How would you define impute

A

Refers to different statistical tools which can be used to calculate missing values from the dataset

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
32
Q

How would you define classification

A

The process of using machine learning to identify different cases based on patterns found in data (example: spam not spam)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
33
Q

What is a common cluster finding model

A

K - means

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
34
Q

How old you define training dataset

A

The data on which the model will be trained

35
Q

What is the result of splitting the dataset

A
  1. A training dataset

2. A test dataset

36
Q

Why do you split the dataset

A

To test against the bias variance - trade off

37
Q

What does splitting the data accomplish

A

Allows you to keep some data hidden during training so that data can be used to evaluate your model before you put it into production

38
Q

Name four unsupervised learning tasks

A
  1. Clustering
  2. Visualization
  3. Dimensionality
  4. Association Rule Learning
39
Q

How much of your data will be in the training dataset after splitting

How much of your data will be in the test dataset after splitting

A
  1. Training will be 80%

2. Test will be 20%

40
Q

How would you define test dataset

A

The data withheld from the model during training which is used to test how well your model will generalize to new data

41
Q

How would you define unsupervised learning

A

There are no labels for the training data, the algorithm tries to learn the underlying patterns or distributions that govern the data

42
Q

Name three common unsupervised learning tasks

A
  1. Clustering
  2. Association
  3. . Dimensionality Reduction
43
Q

What does a transformer do

A

Enables training over larger datasets involving sequences of data

  • it is a more modern replacement for RNN/LSTMs
44
Q

What are the two most common supervised tasks

A
  1. Regression

2. Classification

45
Q

How would you define supervised learning

A
  1. Data is labeled (already has the solution)
  2. Every training sample from the dataset
  3. Has the corresponding label or output
  4. Value associated with it and as a result the algorithm learns to predict labels or output
46
Q

How would you define reinforcement learning

A

The algorithm figures out which actions to take in a situation to maximize a reward (in the form of a number) on the way to reaching a specific goal

47
Q

What is the difference between supervised and unsupervised learning

A

Supervised learning uses labeled input and output data, unsupervised learning does not have/use labeled data

48
Q

What are two main identifies for supervised leaning tasks

A
  1. Categorial label

2. Continuous label

49
Q

How would you define silhouette coefficient

A

A score from -1 to 1 describing the clusters found during modeling

50
Q

How would you define stop words

A

A list of words removed by natural language processing tools when building a dataset

51
Q

In silhouette coefficient what does a score loss than zero indicate

A

Successful identification of discrete non-overlapping clusters

52
Q

Which of the following problem statements fir the definition of a regression - based task

  1. I want to detect when my cat jumps on the dinner table, so I set up a camera and write a program to determine if my cat is in the frame or is not in the frame
  2. I want to determine the expected reading time for online news articles, so I collect data on my reading time for a week and write a browser plug-in to use that data to predict the reading time for new articles
  3. I believe my customers fall into one of many customer segments, but I don’t know what those segments are in advance. After asking for permission, I collect a bunch of data on their actions when they use my reduction and try to determine if there are any collections of users that behave in similar ways
  4. I work for a shore company and want to provide a service to help parents predict their children’s show size for any particular age. Within this system, I represent show size as a continuum of values and then round to the nearest show size
A

2 and 4

53
Q

True or False : Your data requirements will not change based on the machine learning task you are using

A

False

54
Q

True or False : Models are universal, so the data is not relevant

A

False

55
Q

True or False: Data needs to be formatted so that it is compatible with the model and model training algorithm you plan to use

A

True

56
Q

True or False: Data Visualizations are the only way to identify outliers in your data

A

False

57
Q

True or False : After you start using your model (performing inference), you don’t need to check the new data that it receives

A

False

58
Q

True or False : The loss function measures how far the model is from its goal

A

True

59
Q

Why do you need to split the data into training and test data prior to beinginning the model training

A

If you use all the data you have collected during training, you won’t have any with which to test the model during the model evaluation phase

60
Q

What makes hyperparameters different than model parameters

A

Hyperparameters are not updated during model training and are set manually

61
Q

True or False: Model evaluation is not very case - specific

A

False

62
Q

What does model inference involve

A
  1. Generating predictions
  2. finding patterns in your data
  3. using a trained model
  4. testing your model on data it has not seen before
63
Q

True or False : Only a single metric can be used to evaluate a machine learning model

A

False

64
Q

True or False : The model training algorithm updates a model’s parameters to minimize some loss function

A

True

65
Q

True or False : Supervised learning uses labeled data while training a model, and unsupervised learning uses unlabeled data while training a model

A

True

66
Q

In silhouette coefficient what does a score of near zero indicate

A

overlapping clusters

67
Q

What can RNN/LSTMs be used for

A

Processing sequences of data

68
Q

What do RNN/LSTMs do

A

Structured to effectively represent for loops in traditional computing, collecting data while iterating over some object

69
Q

What is a labeled training set

A

a training set that contains the desired solution(aka label) for each instance

70
Q

How would you define the term labeled

A

Data that already contains the solution

71
Q

How would you define plane

A

A mathematical term for a flat surface (like a piece of paper) on which two points can be joined by a straight line

72
Q

How would you define neurons

A

Simple computational units of neural networks

73
Q

How would you define weights

A

Mathematical representations of how much information to allow to flow from one neuron to the next

The trainable model parameters that are the connections between neural networks

74
Q

Name four neural network structures

A
  1. FNNN
  2. CNN
  3. RNN/LSTM
  4. Transformer
75
Q

How would you define neural networks

A

Acollection of very simple models connected together

76
Q

What is good tool to use to see outliers and trends in your data

A

Data Visualization

77
Q

How would you define outliers

A

Data points that are significantly different from others in the same sample

78
Q

How would you define model parameters

A

Settings or configurations the training algorithm can update to change how well the model behaves

79
Q

What are some other terms for model parameters

A
  1. Weights

2. Biases

80
Q

What does a model training algorithm actually do

A

Iteratively update model parameters to minimize some loss function

81
Q

How are model training algorithms used to train a model

A

They work through an iterative process where the current model iteration is analyze to determine what changes can be made to get closer to the goal. Those changes are made and the iteration continues until the model is evaluated to meet the goals

82
Q

What is the first step in model training

A

Randomly split the data set

83
Q

How would you define model inference

A

When the trained model is used to generate predictions

Introduction to Machine Learning (2 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions