Machine Learning And It's Types

                          

                          Machine Learning And It's Types

Machine learning is a subset of artificial intelligence(AI). Machine learning is growing very fastly because it's a new technology in the market and due to a large scale industry involvement it becomes trending technology.

Machine learning is a way to automate the task without being explicitly programmed in the system.

It is possible due to self-learning techniques of machines through algorithms which already feed into it. So, the algorithm is a key factor for machine learning because better the algorithm better will be predictions on datasets and result in a well-defined output.

Let's talk about types of Machine learning and the algorithms for those types:

1. Supervised Learning

2. Unsupervised Learning

3. Semisupervised Learning

4. Reinforcement Learning

(A) Supervised Learning

In supervised learning, data points have known outcome. It is the task of inferring a function from labeled training data. By fitting to the labeled training set, we want to find the most optimal model parameters to predict unknown labels on the test set. If the label is from the limited number of values where these values are unordered,(i.e. output is the category)  then it's a classification problem otherwise if the label is a continuous value output or real number, then it is a regression problem.

                                              

                                               fig:-(a) Regression(continues-valued output)

            

                                                        

                                                     fig:-(b) Classification of leaf( 3-different species)

(B) Unsupervised Learning

 In Unsupervised learning, data points have an unknown outcome that is we don't know about the output i.e. train set is unlabeled. It's possible to observe some similarities between groups of objects and include them in appropriate clusters. 

                           fig:-(c) Unsupervised problem(prediction of different fruits from the sample)

(C)  Semi-supervised Learning

  Semi-supervised is a class of supervised learning tasks and techniques and uses both labeled and unlabeled data for prediction. The method allows us to significantly improve accuracy because we can unlabeled data in the train set with a small amount of labeled data. 

    (D) Reinforcement Learning

  It is an important type of machine learning in which an agent learn how to behave in an environment by performing actions and seeing the results. Reinforcement learning is said to be hope of true artificial intelligence.

                                                     fig: Reinforcement Learning

So, these are types of machine learning, if you want to study ml you are going through these types and understand it more.
I hope you like this article on machine learning.Feedback and suggestions in comment sections are more appreciable.