Must Read Types of Machine learning

As wikipedia says, Machine learning (ML) is the study of computer algorithms that improve automatically through experience. It is seen as a subset of artificial intelligence. Machine learning algorithms build a model based on sample data, known as “training data”, in order to make predictions or decisions without being explicitly programmed to do so.

So, let’s simplify the defination in simple words.We humans learn from our past experiences, and machines follow instructions given by human.What if we humans train machines to learn from past data and faster than humans as well, that’s called machine learning but it’s a lot more than just learning, it’s also about understanding and reasoning.

For Example:Say, you provide a system with the input of data that contain different photographs of human.Now, you want the system to figure out and group this photographs into male and female group.Now what the system does, it analyses the data then it tries to find patterns, shape, colour, hairs, based on them the system will try to predict the different humans and group them.Finally the system will keep track its all decisions in the process to make sure its learning.The next time you ask the system to predict and segregate the different humans from the photographs it won’t have to go through the entire process again.This is how Machine Learning works.

Types of Machine Learning:

Image for post
Image for post

Supervised Learning

(the system learns from the training data that is labeled)

In order to understand Supervised learning, first let’s understand two terms.

What is training data? Algorithms learn from data. They find relationships, develop understanding, make decisions, and evaluate their confidence from the training data they’re given. And the better the training data is, the better the model performs.

  • Proper Definition is: Neural networks and other artificial intelligence programs require an initial set of data, called training data, to act as a baseline for further application and utilization. This data is the foundation for the program’s growing library of information.

What is a test set?

  • Once a model is trained on a training set, it’s usually evaluated on a test set. Oftentimes, these sets are taken from the same overall dataset, though the training set should be labeled or enriched to increase an algorithm’s confidence and accuracy.
Image for post
Image for post

In Supervised Learning, the dataset on which we train our model is labeled. There is a clear and distinct mapping of input and output.We train our machines with some training data and test them used test data, during this process, it will remember its decisions and next time when we provide them data, It will give us the result based on its previous training or learning.

Unsupervised Learning

(Non-labeled training data)

In simple words, unsupervised learning is when no labels or information is provided with respect to the data set that has been provided.

In unsupervised learning model, the training dataset isn’t labelled, in contrast, the algorithm reads through the non-labelled data and uses ‘clustering analysis’ to discover classes within the data. The algorithm sorts the data and uses this data to predict the output for the ‘test data’.

Image for post
Image for post
Source: Google

Let’s see some differences in Supervised and Unsupervised Learning:

  1. The data is labeled in Supervised, but In unsupervised we have Non-labeled data.
  2. In supervised learning you get a feedback.
  3. Supervised Learning is used to predict data whereas unsupervised learning is used to predict patterns or structures in data.

Reinforcement Learning

(The machine learns on its own.)

Image for post
Image for post

In Reinforcement Learning, system first provided with the data, to provide us some result, When system provides the data, then we tell system it’s wrong and now it’s the right answer. Basically we provide feedback to the system, now the system will learn the correct data provided by us.Next time when we ask system,it will provide us with the correct output.

Reinforcement learning model uses the goal-oriented algorithms. It is used in various machines in order to define the best possible behavior to be taken in a specific situation. The major difference between supervised and reinforcement learning is that supervised learning model uses the training dataset that has an answer key, the labels and classes. Hence the model is trained with the correct answer itself whereas, in reinforcement learning, there is no answer key, the agent has to decide how to perform in a given situation. It is all about taking specific action in a particular situation in order to maximise the reward.

Thanks for your valuable time.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store