Have you ever wondered how when you look for a wireless mouse on Amazon you get suggestions from Amazon to buy other products related to wireless mice? or how is Facebook so intelligent that it recommends who can be your friend? All this is possible because of Machine Learning. Today, I am going to discuss what Machine Learning is and what are different types of Machine learning. Machine learning is becoming ubiquitous, more and more companies are deploying machine learning models in their systems and there are so many people who want to know what it actually is and are confused that why the terms Artificial Intelligence and Machine learning are used interchangeably.
Machine learning is a core subfield of Artificial Intelligence (AI). In Machine learning, we deploy statistical methods to make programs that learn from data. we hope to develop a model which can adapt to the environment without needing to code an explicit set of rules. The machine learns with experience and gets better with experience.
According to one of the founders of machine learning, Professor Tom Mitchell, Machine learning is about how we create computer programs that improve with experience. We say that a computer program is capable of learning if its performance in doing task T improves with more experience E, where the performance is measured by some performance measure or evaluation P. And we allow the computer to gain more experience E with more data. With more experience, our computer program will hopefully be able to learn patterns and do a better job at doing its tasks.
We divide the data between training and testing data. Training data is used to train the machine learning model (which is a mathematical function) and testing data is used to validate the model.
Supervised Learning
Supervised Learning is a machine learning technique in which our training data is labeled, which means we feed the model with the corresponding outputs for all our input. The model then uses the Regression or Classification model to make predictions for the testing data.
Unsupervised Learning
As opposed to supervised Learning, in unsupervised learning training data is not labeled, which means we don’t feed the model with the corresponding outputs for our input data. The model then makes the cluster and makes predictions for the testing data.
Deep Learning
Deep Learning has supervised learning in which an artificial neural network algorithm is used.
Reinforcement Learning
In Reinforcement learning, the machine learns from the environment and gets better by getting positive and negative rewards from the environment.
In Machine Learning, there are different algorithms that can be used for different tasks at hand. More about these algorithms used in machine