ai Flashcards

1
Q

What is clustering?

A

Clustering is an unsupervised learning technique that groups unlabelled data into clusters/categories of similar data.

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

What is a class in supervised learning?

A

A class is a category given to a supervised learning model for it to classify given data as a particular type of input.

e.g. a class for cats, a class for dogs

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

What is a reward in reinforcement learning?

A

The reward is the positive or negative feedback (e.g. a score) given to an AI after a set of actions during reinforcement learning.

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

What is intelligence?

A

Intelligence is the ability to apply prior knowledge/skills/experience to a scenario effectively.

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

What does artificial mean?

A

Artificial refers to something made by or produced by human beings, rather than occurring naturally.

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

What is Artificial Intelligence (AI)?

A

Artificial Intelligence (AI) is a human-made system that can simulate human intelligence, e.g. the ability to learn, reason, and apply knowledge/skills to new scenarios.

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

What is machine learning?

A

Machine learning is a type of Artificial Intelligence that learns by using large sets of data, instead of following step-by-step instructions.

e.g. voice recognition software, identifying objects in images, playing complex games.

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

What are the three areas of machine learning?

A

The three areas of machine learning are supervised learning, reinforcement learning, and unsupervised learning.

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

What is unsupervised learning?

A

Unsupervised learning is a type of machine learning used to analyze a set of unlabelled data to find hidden patterns/groups.

e.g. grouping customers by purchasing patterns, detecting patterns in internet usage, detecting weather patterns in sensor data.

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

What is unlabelled data?

A

Unlabelled data is data where its context is not defined, meaning no information is given about the data to be used. It is used in unsupervised learning.

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

What is association in unsupervised learning?

A

Association is an unsupervised learning technique that finds relationships between unlabelled data to discover trends and patterns.

e.g. used to recommend new purchases or new videos to a user based on large datasets of existing data.

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

What is supervised learning?

A

Supervised learning is a type of machine learning where the AI model is trained using example (labelled) data to model the desired output.

e.g. training an AI to recognize objects within images.

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

What is labelled data?

A

Labelled data is data where one or more labels are given to provide information about what is contained.

e.g. images of birds labelled with the type of bird contained in the image.

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

What is reinforcement learning?

A

Reinforcement learning is a type of machine learning without initial training data, where an AI model learns through trial and error with feedback on its output.

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

What is bias in AI?

A

Bias in AI occurs when a particular result/output is favoured over another. AI is not inherently biased but can become biased based on the data it is trained on.

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

What are the causes of AI bias?

A

Causes of AI bias include:
- The data collected does not fully represent the situation being modeled.
- The dataset to train the model is not large enough.
- Bias in society resulting in bias when collecting data for the model.