Skip to main content

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.


                                 

Comments

Post a Comment

Popular posts from this blog

Rising of the AI in the human centric Development

Rising of the AI in the human centric Development The rising of the artificial intelligence in later 90's have make a rapid impact in field of technology and from 21st century the blooming of a mechanism makes several impact in various industries including software, education, healthcare and many more. As the world becomes increasingly reliant on technology, the role of  artificial intelligence  (AI) in human-centric development has risen to the forefront. From healthcare to transportation to education, AI is being used to improve the lives of people around the globe. Major areas where artificial intelligence AI makes an Impact One area where AI has made significant strides is in the healthcare industry. AI-powered virtual assistants can now assist doctors in diagnosing and treating patients, freeing up valuable time for medical professionals. In addition, AI-powered wearable devices can track a person's health and alert them to any potential issues. The transportation industr...
Post a Comment
Read more

How to calculate Running Time of an algorithm

                                            Calculate Running Time of an Algorithm The running time of algorithm defines the time required to execute an algorithm on the given set of inputs(n). There are mainly three types of complexity cases defines to measure the running time of an algorithm also known as Asymptotic analysis. 1) Best Case : Best case also called ( Ω) omega  notation which measure the best case scenario of how long an algorithm can possible take to complete given operation on (n) inputs. It's also known as lower bound. 2) Average Case : It represents by ( Θ) theta  notation which measure the average time requires to complete a given operation on set of inputs. It measures between upper and lower bound running time and calculate average running time. 3) Worst Case: It defines the worst case running time of an algorithm. Also represent using ( Ο) Big-o...
Post a Comment
Read more

Machine Learning and It's Types

                           Machine Learning and It's Types                                 Machine Learning is ability to automatically learn and improve from experience without being explicitly programmed. So rather than typing the code for all the times and do knowledge engineering, machine learning helps the machine  to learn from previous data and find insights and pattern from it.  Basically Data is train on given data set and and applied machine learning algorithm and it find insights. Simply put, Machine learning makes a computer act and think like a human. Types of machine learning           Supervised Learning In supervised learning you use labeled data,which is a data set that has been classified, to infer a learning algorithm. The data set is u...
2 comments
Read more