Brainscape's Knowledge GenomeTM

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


See full index

AI-900 > part 1 > Flashcards

part 1 Flashcards

1
Q

AI define

A

Machines performing jobs mimicking human behaviour

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

ML define

A

Foundation of an AI system, learns and predicts like a human
Machines that get better without explicit programming

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

DL define

A

Machines that have an artificial NN inspired by the human brain to solve complex problems

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

Data scientist define

A

Person with multi-disciplinary skills in maths, stats, predictive modelling and ML to make future predictions

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

Describe onion diagram of AI, ML, DL

A

AI contains ML which contains DL

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

Anomoly detection

A

Detects outliers or things out of place like a human

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

Computer vision

A

be able to see like a human

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

NLP

A

Be able to process human languages and infer context

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

Conversational AI

A

be able to hold a conversation with a human

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

What is a dataset

A

Logical grouping of units of data that are closely related and/or share the same data structure

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

MNIST

A

Images of handwritten digits used to test classification, clustering and image processing algorithms e.g. computer vision ML models

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

COCO (common objects in context) dataset

A

Contains common images using a JSON file (coco format) that identify objects or segments within an image
- features object segmentation, recognision in context, superpixel stuff segmentation
Azure has a data labelling service which can export in coco format

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

Data labeling

A

Identifying raw data and adding one more more meaningful and informative labels to provide context so ML model can learn

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

data labelling - supervised

A

Labels are a prerequisite to produce training data. Each piece generally labelled by a human

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

data labelling - unsupervised

A

Labels produced by machine, might not be human readable

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

ground truth

A

Properly labelled dataset used as objective standard to train and assess the model. Accuracy of trained model depends on accuracy of ground truth

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

Supervised learning

A

Data that has been labelled for training.
Task-driven - make a prediction
When the labels are known and you want a precise outcome. You need a specific value returned e.g. Classification, Regression

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

Unsupervised learning

A

Data has not been labelled, ML model needs to do its own labelling
Data-driven - recognise a structure or pattern
When labels not known and outcome doesn’t need to be precise.
Trying to make sense of data.
e.g. Clustering, dimensionality reduction, association

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

Reinforcement learning

A

No data, there is an environment and an ML model generates data any many times to reach a goal
Decisions-driven - Game AI, Learning Tasks, Robot Navigation

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

Neural network

A

Mimicking the brain. Node/neuron represents an algorithm
Data inputted into neuron and based on output, data passed to one of many other connected neurons.
Connections are weighted.
Network is organised into layers
Input layer, many hidden, and an output

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

How many layers for a NN to be called deep learning

A

3+

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

Feed Forward (FNN)

A

Neural networks where connections between nodes don’t form a cycle (always moving forward)

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

Back propagation

A

Moves backwards through the neural network adjusting weights to improve next iteration’s performance. How the Neural net learns.

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

Loss function

A

Function comparing ground truth to prediction to determine error rate. Performs calculation at the end, performs calculation and then back propagates.

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

Activation functions

A

Algorithm applied to a hidden layer node that affects connected output (e.g. ReLu, part of backpropagation)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

Dense

A

When the next layer increases the number of nodes

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

Sparse

A

When the next layer decreases the amount of nodes
(dimensionality reduction is when nodes decrease from one layer to the next)

28
Q

GPU

A

General processing unit specifically designed to render high res images and video concurrently
Can perform parallel operations on multiple sets of data - used for non-graph tasks e.g. ML and scientific computation

29
Q

CPU cores vs GPU cores

A

CPU - 4-16 processor cores on average
GPU - thousands of processor cores

30
Q

CUDA (compute unified device architecture)

A

Parallel computing platform and API by NVIDIA allowing developers to use CUDA-enabled GPUs for general purpose computing on GPUs (GPGPU)

31
Q

NVIDIA

A

Company manufactures GPUs for gaming and professional markets

32
Q

Major deep learning frameworks are integrated with

A

NVIDIA deep learning SDK - collection of NVIDIA libraries for deep learning, e.g. cuDNN (CUDA deep neural network library)

33
Q

ML Pipeline stages

A

Data labelling -> supervised learning so model can learn by example
Feature engineering -> translate data to format ML models can understand
Training -> multiple iterations, getting smarter
Hyperparameter tuning -> Try different parameters to optimise outcome
Serving -> So model is accessible, host in VM or container
Inference -> requesting to make prediction e.g. real time endpoint (for one request), or batch processing (slower, but could also be real time)

34
Q

Forecasting

A

Future prediction with relevant data: analysis of trends, not ‘guessing’

35
Q

Prediction

A

Make future prediction without relevant data: uses statistics to predict future outcomes, more ‘guessing’, uses decision theory

36
Q

Performance/Evaluation metrics

A

Used to evaluate ML algorithms:

37
Q

Different types of metrics for different problems

A
  • Classification metrics (accuracy, precision, recall, F1-score, ROC, AUC)
  • Regression metrics (MSE, RMSE, MAE)
  • Ranking metrics
  • Statistical metrics (correlation)
  • Computer vision metrics
38
Q

Jupyter Lab

A

Will replace Jupyter notebook

39
Q

Regression

A

Process of finding a function to correlate a labelled dataset (supervised) into continuous variable/number e.g. what will temperature be

40
Q

Regression error

A

Distance of vector from regression line. Used to predict future variables - MSE, RMSE, MAE

41
Q

Classification

A

Finding a function to divide a labelled dataset into classes/categories e.g. what weather category will it be. (supervised)

42
Q

Classification line

A

Divides dataset with one side being one category, another being another category

43
Q

Classification algorithms

A

Logistic regression, decision tree/random forest, neural networks, naive bayes, k nearest neighbours, SVM

44
Q

Clustering

A

Process of grouping unlabelled data based on similarities/differences (unsupervised)
e.g. K-means, K-medoids, Density based, Hierarchichal

45
Q

Confusion Matrix

A

Visualise model predictions vs ground truth labels (actual). Aka error matrix.
Top labels: predicted no, predicted yes
Side labels: actual no, actual yes

46
Q

Size of confusion matrix

A

Number of categories x 2 (ground truth x prediction)

47
Q

use cases for anomaly detection

A

data cleaning
intrusion detection + fraud detection
systems health monitoring
sensor networks event detection
ecosystem disturbances

ML is more accurate than by hand and more efficient + accurate

48
Q

computer vision DL algorithms

A

CNN - image + video recognition inspired by how eyes process info + send to brain
Recurrent NN (RNN) - handwriting/speech recognition

49
Q

Types of computer vision

A

image classification
object detection
semantic segmentation (identify segments + objects by drawing pixel mask) - good for objects in movement
image analysis - analyse image/video to apply descriptive + context labels
optical character recognition
facial detections

50
Q

Azure Computer Vision. iOS app built

A

Seeing AI developed for iOS, use device camera to identify people + objects + device audibly describes for visually impaired

51
Q

Azure computer vision service offering

A

‘Computer Vision’ - analyse image/videos + extract description, tags, objects, text
‘Custom vision’ - Custom image classification + object detection models using own images
Face - Detect + identify people and emotions in images
Form recogniser - translate scanned docs into key/val or tabular editable data

52
Q

NLP

A

ML that understands context of a corpus enabling
- analyse/interpret text in docs/emails
- interpret + contextualise spoken token e.g. sentiment analysis
- synthesise speech
- automatically translate
- interpret spoken or written commands + determine appropriate actions

53
Q

NLP Azure service offering

A

Text analytics - sentiment analysis, key phrase extraction, identify language, entity recognition
Translator - real-time translation
Speech - transcribe into searchable text
LUIS (Language understanding) - NLP enabling understanding human language in own application

54
Q

Conversational AI

A

Tech that can participate in conversations w humans: chatbots, voice assistants, Interactive Voice Recognition Systems

55
Q

Conversational AI use cases

A

Online Customer Supports
Accessibility e.g. visually impaired
HR Processes - employee training
Healthcare
IoT
Software e.g. autocomplete search

56
Q

Conversational AI Azure services

A

QnA Maker - create conversational q and a bot from knowledge base
Azure Bot Service - deploys the bot created with QnA maker. Intelligent serverless bot service scaling on demand. For creating/publishing/managing bots.

57
Q

Responsible AI

A

ethical, transparent + accountable use of AI

58
Q

Microsoft AI principles

A

Fairness
Reliability + Safety
Privacy + Security
Inclusiveness
Transparency
Accountability

59
Q

Principle: Fairness

A

AI systems should treat all people fairly
Bias can be introduced during pipeline development, reinforcing societal stereotypes
E.g. systems dealing w opportunities/resources/info in criminal justice/employment/finance
Azure ML can tell you how each feature can influence model’s prediction for bias

60
Q

Principle: Reliability + Safety

A

Should perform reliably + safely. Rigorous testing needed to ensure works as expected before end user release + shortcomings reported to user.
Critical safety importance: Autonomous vehicle, AI health diagnosis, autonomous weapons

61
Q

Principle: Privacy + Security

A

Nature of ML model may require personally identifiable information
Ensure data protected so no leaking/disclosing.
Some cases, model can be run on user’s device avoiding vulnerability

62
Q

Principle Privacy + Security: AI security principles to detect malicious actors

A

data origin and lineage, data use internal vs external, data corruption considerations, anomaly detection

63
Q

Principle: Inclusiveness

A

Design AI solutions for minority then can design AI for majority e.g. physical ability, gender, sexual orientation, ethnicity etc

64
Q

Principle: Transparency

A

AI systems should be understandable. Interpretability/intelligibility is when end users can understand UI behaviour.
Transparency: mitigates unfairness, helps debug systems, gains user trust
Open about why using AI+ limitations
Open source AI framework can help

65
Q

Principle: Accountability

A

Structure put in place enacting AI principles + putting into account.
AI should work within framework of governance + organisational principles, ethical + legal standards clearly defined.

Principles guide Microsoft on how they develop, sell + advocate when working w 3rd parties, pushing towards regulations towards AI principles

AI-900 (2 decks)

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