Brainscape's Knowledge GenomeTM

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


See full index

a-eye-nine-humdred > 02-Machine Learning > Flashcards

02-Machine Learning Flashcards

1
Q

How do machines learn

A

Machines learn from data

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

What is Regression

A

Regression is a form of Machine Learning that predicts a numeric LABEL based on an item’s FEATURES

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

What type of Machine Learning technique is Regression

A

Regression is Supervised Machine Learning

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

What is Supervised Learning

A

Technique in which you train a model using data that includes both FEATURES and known values for the LABEL, so that the model learns to FIT the FEATURE combinations to the LABEL

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

What is Classification

A

Classification is a form of Machine Learning that predicts which category, or class an item belongs to

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

What type of Machine Learning technique is Classification

A

Classification is Supervised Machine Learning

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

What is Clustering

A

Clustering is a form of Machine Learning that is used to GROUP SIMILAR ITEMS based on their features

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

What type of Machine Learning technique is Clustering

A

Clustering is Unsupervised Machine Learning

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

What is Unsupervised Machine Learning

A

Unsupervised Machine Learning is where you train a model to separate items into clusters based purely on their characteristics, or FEATURES. There is no previously known cluster value (or LABEL) from which to train the model.

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

What are Azure Machine Learning services

A
  1. AUTOMATED machine learning
  2. Azure Machine Learning DESIGNER
  3. Data and compute MANAGEMENT
  4. PIPELINES
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is Automated Machine Learning

A

Automate Machine Learning ALLOWS NON-EXPERTS to create an effective machine learning model from data very quick

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

What is Azure Machine Learning designer

A

Azure Machine Learning designer is a GUI that allows no-code development of machine learning solutions.

The Designer tool in Azure Machine Learning studio allows you to create and run pipelines by using DRAG & DROP INTERFACE to connect modules that define the steps and data flow for the pipeline.

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

What is Data and Compute management

A

Data and Compute management is CLOUD-BASED DATA STORAGE AND COMPUTE resources that professional data scientists can use to run data experiment code at scale.

Scale meaning they can run multiple training experiments in parallel while incurring costs only when actually used.

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

What are Pipelines

A

Pipelines are MULTI-STEP WORKFLOWS to
PREPARE data,
TRAIN models,
and perform model MANAGEMENT tasks.

Pipelines allow data scientists, software engineers, and IT operations professionals to do the above.

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

What is Forecasting

A

Forecasting is Regression with a TIME-SERIES element

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is Azure Machine Learning

A

Azure Machine Learning is a CLOUD SERVICE that you can use to TRAIN and MANAGE machine learning models.

You need COMPUTE on which to run the training process.

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

What is a Workspace

A

Workspace is CREATED IN AZURE SUBSCRIPTION to use Azure Machine Learning.

It allows you to MANAGE data, compute resources, code, models, and other ARTIFACTS related to your machine learning workloads

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

What are Compute Targets

A

Compute Targets are cloud-based resources on which you can run MODEL TRAINING and DATA EXPLORATION processes.

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

What are four types of Compute Resources you can create

A
  1. Compute INSTANCES
  2. Compute CLUSTERS
  3. INFERENCE Clusters
  4. ATTACHED Compute
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

What are Compute Instances

A

Compute Instances are DEVELOPMENT WORKSTATIONS that data scientists can use to work with data and models

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

What are Compute Clusters

A

Compute Clusters are scalable CLUSTERS OF VIRTUAL MACHINES for on-demand processing of experiment code

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

What are Inference Clusters

A

Inference Clusters are DEPLOYMENT TARGETS FOR PREDICTIVE SERVICES that use your trained models

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

What is Attached Compute

A

Attached Compute LINKS TO EXISTING AZURE COMPUTE RESOURCES, such as Virtual Machines or Azure Databricks clusters

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

What is a dataset

A

Dataset is an object that ENCAPSULATES DATA for model training and other operations

25
Q

What are experiments

A

Experiments are OPERATIONS you run in Azure Machine Learning

26
Q

What is cross-validation

A

Cross-validation ITERATIVELY tests the trained model with data it wasn’t trained with and compare the predicted value with the actual known value.

27
Q

What are residuals

A

Residuals are the difference between the PREDICTED and ACTUAL value.

Amount of ERROR in the model.

This particular performance metric is calculated by SQUARING the errors across all the test cases, finding the MEAN of these squares, and then taking the SQUARE ROOT.

The smaller the value, the more accurate the model is at predicting.

28
Q

Which are two ways you can deploy a Machine Learning model as a service

A
  1. Azure Container Instance (ACI)

2. Azure Kubernetes Service (AKS)

29
Q

What is Reinforcement Learning

A

In Reinforcement learning, the algorithm CHOSE AN ACTION IN RESPONSE TO EACH DATA POINT.

Common in robotics where set of sensor readings at one point in time is a data point, and the algorithm must CHOOSE THE ROBOT’S NEXT ACTION. Also for Internet of Things.

Algorithm also RECEIVES A REWARD SIGNAL a short time later, indicating how good the decision was.

Based on this signal, the algorithm modifies its strategy in order to achieve the highest reward.

30
Q

What type of ML algorithm predicts values

A

Regression algorithms predict values. Makes forecasts by estimating the relationship between values. Answers questions like: How much or how many?

31
Q

What ML algorithm finds unusual occurrences

A

Anomaly Detection algorithm finds unusual occurrences. Identifies and predicts rare or unusual data points

32
Q

What ML algorithm discovers structure

A

Clustering algorithm. It separates similar data points into INTUITIVE GROUPS. Answers questions like: How is this organized?

33
Q

What ML algorithm generates recommendations

A

Recommenders algorithm predicts what someone will be interested in. Answers the question: What will they be interested in?

34
Q

What is K-Means

A

K-Means is a Clustering algorithm, unsupervised learning

35
Q

What ML algorithm classifies images

A

Image Classification is ML algorithm that classifies images.

Classifies images with popular networks.

Answers questions like: What does this image represent?

36
Q

What is a Confusion Matrix

A

TABULATION of the predicted and actual value counts for each possible class

37
Q

What are true positives

A

Both predicted and actual values are TRUE

38
Q

What are true negatives

A

Predicted and actual values are FALSE

39
Q

What are false negatives

A

Predicted value is FALSE but actual value is TRUE

40
Q

What are false positives

A

Predicted value is TRUE but actual value is FALSE

41
Q

What is accuracy

A

Ratio of CORRECT predictions (true positive + true negative) to the TOTAL number of predictions

42
Q

What is precision

A

True Positives / (True Positives + False Positives). Of all the ones that are PREDICTED positives, which ones were correctly predicted.

43
Q

What is recall

A

True Positives / (True Positives + False Negatives). Of all the ones that are ACTUAL positive, which ones were correctly predicted.

44
Q

What is F1 Score

A

An overall metric that combines PRECISION and RECALL

45
Q

What is ROC curve

A

Received Operator Characteristic.

Plot TRUE POSITIVE rate against FALSE POSITIVE rate. The larger area under the curve, the better the model is performing.

46
Q

What is AUC

A

Area under the curve. Metric for binary classification.

47
Q

What are centriods

A

Centroids are randomly initialized K COORDINATES

48
Q

What is average distance to other center

A

This indicates HOW CLOSE, on average, EACH POINT in the cluster is TO THE CENTRIOD of all other clusters

49
Q

What is average distance to cluster center

A

This indicates how close, on average, each point in the cluster is to the centroid of the cluster

50
Q

What is number of points

A

The number of points assigned to the cluster

51
Q

What is maximal distance to cluster center

A

The maximum of the distances between each POINT and the CENTROID of the point’s cluster.

If this number is high, the cluster may be widely dispersed.

This statistic in combination with the Average Distance to the Cluster Center helps you determine the cluster’s spread.

52
Q

What question does Anomaly Detection answer

A

Is this weird?

53
Q

What question does Image Classification answer

A

What dose this image represent?

54
Q

What question does Two-Class Classification answer

A

Is this A or B?

55
Q

What question does Multiclass Classification answer

A

Is this A or B or C or D?

56
Q

What question does Recommenders answer

A

What will they be interested in?

57
Q

What question does Text Analytics answer

A

What info is in this text?

58
Q

What question does Regression answer

A

How much or how many?

a-eye-nine-humdred (5 decks)

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