ML TYPES

Question 1

What is supervised learning?

Answer 1

Here the machine is given training examples of inputs and their co-responding outputs so that when new data is given it can predict the new output.

Question 2

How to train data in supervised?

Answer 2

The model gets trained on ‘labelled dataset’. Labelled dataset has both input and output parameters.

Question 3

How does this training help?

Answer 3

We train the system with the data which is well labeled with correct output.

As the training period progresses, the algorithm is able to identify the relation between two variables. With this it can predict new output

Question 4

What are types of supervised learning?

Answer 4

Classification and Regression

Question 5

What is classification?

Answer 5

Predicts the output in form of category

Question 6

What is the output variable in classification?

Answer 6

category

Question 7

What are types of classification?

Answer 7

binary, multi-class

Question 8

what is binary?

Answer 8

If the algorithm labels the input into two categories then it is binary classification

Question 9

what is multi-class

Answer 9

If the inputs are sorted into multiple classes then it is multi-class classification

Question 10

what is regression?

Answer 10

Aregression modelpredicts a numeric value.

Question 11

what is the output variable in regression?

Answer 11

The output variable is a real value
The output variable is a single value

Question 12

what is un-supervised learning?

Answer 12

Deals with the unlabeled data (photos, videos, audio etc.). No training dataset is provided to the the machine i.e., machine works on its own to discover information.

Question 13

what is the goal of un-supervised learning?

Answer 13

An unsupervised learning model’s goal is to identify meaningful patterns among the data.

Question 14

what is the goal of supervised learning?

Answer 14

The goal is to produce an accurate enough mapping system that when new input is given the system can predict the correct output.

Question 15

is supervision required in unsupervised?

Answer 15

no

Question 16

what does unsupervised learning work with?

Answer 16

unlabeled data and uncategorized data

Question 17

give working of unsupervised learning?

Answer 17

1. unlabeled data is given
   2.fed to machine
   3.model will interpret raw data to find hidden patterns
   4.then is divides the obj into groups based on similarities and differences

Question 18

how is data sorted in unsupervised?

Answer 18

Sorts data on the basis of similarities, differences and patterns.

Question 19

what are types of un-supervised?

Answer 19

clustering and association

Question 20

what is clustering?

Answer 20

grouping the objects into clusters such that objects with most similarities remains into a group and has less or no similarities with the objects of another group.

Question 21

what is association?

Answer 21

finding the relationships between variables in the large database. It determines the set of items that occurs together in the dataset.

people who buy X item (suppose a bread) are also tend to purchase Y (Butter/Jam) item.

Question 22

which is preferred?

Answer 22

unsupervised?

Question 23

why is unsupervised preferred?

Answer 23

Unsupervised learning is preferable as it is easy to get unlabeled data in comparison to labeled data.

Question 24

what is reinforcement learning?

Answer 24

Computer learns to perform a task through repeated trial and error interactions with a dynamic environment.

Designed to learn from continuous experience rather than data.

Rewards system.

Question 25

what kind of system used in reinforcement learning?

Answer 25

rewards/penalty system

Question 26

what method is used in reinforcement learning?

Answer 26

hit/trial

Question 27

how does reinforcement learn?

Answer 27

through experience

Question 28

what is artificial intelligence?

Answer 28

the process by which machines are able to mimic human behaviour through various algos.

its components are ML and DL.

it exhibits its intelligence using decision making

increase chance of success not caring about accuracy

efficiency = AI+ML

Question 29

what is ML?

Answer 29

study method using statistical methods to improve with experience

subset of AL

exhibits intelligence when system learns from the data

increase accuracy not caring about success

efficiency < DL

Question 30

what is DL?

Answer 30

study done using neural networks to imitate human learning

subset of ML

exhibits intelligence when deep neural networks analyse data and provide the necessary output

increases accuracy most out of the three

efficiency >ML

Question 31

on what basis supervised learning models give correct output?

Answer 31

on the basis of the correct dataset provided

Question 32

how does unsupervised model make predictions?

Answer 32

by giving it data that does not contain correct data or a unlabeled dataset

Question 33

what is generative AI

Answer 33

creates content from user input

can take a variety of inputs and create a variety of outputs, like text, images, audio, and video.