ML paradigms

Question 1

What are the machine learning paradigms?

Answer 1

The paradigms are: Supervised learning, Unsupervised learning, Semi Supervised learning and Renforcement learning.

Question 2

What is machine learning

Answer 2

It gives the computers to learn without being explicitly programmed.

OR

A computer program is said to learn from experience E, for some tasks T, and performance measure, P.

Question 3

What is the difference between traditional programming and machine learning?

Answer 3

Traditional programming:
Input: data, program
and the computer provides the corresponding output.

Machine Learning:
input: data, past output
and the computer learns from the given inputs to generate the most accurate output.

Question 4

Explain the Supervised Machine learning Algorithm

Answer 4

Uses labeled training datasets to predict new, unseen datasets. It is classified into two categories: Classification and Regression.

Question 5

Classification?

Answer 5

It classifies the given datas in categories. it is used for discrete type of outputs.
eg: classification of different types of fruits.

Question 6

Regression?

Answer 6

used in real-world data like predicting a real value. The output is in the form of continuous values.
eg: house price prediction, mark prediction, etc.

Question 7

Supervised learning is also known as?

Answer 7

Inductive Learning

Question 8

Steps involving in supervised learning?

Answer 8

1. Data collection
2. dividing the data into training and testing data
3. Label the training sets
4. Extracting the features
5. Training the training set
6. Make predictions on the unseen data

Question 9

Explain Unsupervised learning?

Answer 9

It is a machine learning algorithm that is used to draw inferences from unlabelled datasets.

The most common method in unsupervised learning is clustering and association.

Question 10

What is clustering?

Answer 10

A method of grouping the objects into clusters such that the data with most similarities remains into a group and the others with least similarity remain in other groups.

Question 11

What is association?

Answer 11

It is used for finding relationships between the variables in a large database. it is mainly used in business fields.

Question 12

Explain semi-supervised learning?

Answer 12

An intermediate group between supervised and unsupervised learning algorithms.
Uses a combination of labelled and unlabelled datasets during the training period.

Question 13

Explain the working of Semi-Supervised Learning?

Answer 13

1. It trains the model with less amount of training data. The training continues until the model gives accurate results.
2. Now use the unlabelled datasets with pseudo labels and the results may not be accurate.
3. The labels from the labeled training data and the pseudo labels are linked together.
4. The input data in the labelled and unlabelled training data are also linked together.
5. Again train the model with the combined inputs.

Question 14

Explain Renforcement learning algorithm?

Answer 14

Here the agent learns by interacting with its environment.

The agent receives rewards for performing correctly and penalties for performing incorrectly