Supervised, Unsupervised and Reinforcement Machine Learning
Artificial Intelligence has changed the entire 21st century in terms of technology. Machine learning is a part of AI. Machine learning is the study of computer algorithms that improve automatically through experience and by the use of data. In this blog, we’ll be discussing the different types of machine learning and we will compare them with each other.
Machine Learning is the science of getting computers to act by feeding them data and letting them learn a few tricks on their own without being explicitly programmed. We can compare machine learning with a human child. A child will not distinguish between vegetables such as pumpkin, carrots, and onion. But day by day the child collects more information and start developing the capability to distinguish between various vegetables.Our brain has the capacity to think and make decisions since we have been feeding it a lot of data and this is exactly how machine learning works.
There are three types of machine learning
1. Supervised Learning
2. Unsupervised Learning
3. Reinforcement Learning
Supervised Learning
In supervised learning, machine learns under guidance. In here machines learn by feeding them label data and explicitly telling them the input and how the output should be. In simple terms, supervised Learning is a method in which we teach the machine learning using labeled data.
There are 2 main categories of problems in supervised learning. They are regression and classification. Classification is about predicting a label or a class whereas regression is about predicting a continuous quantity. Let us assume that we have to categorize our emails based on spam and not spam emails. In these cases we can use classification algorithms. Let us assume another example to predict the weight of a person. Here possibilities for weight are limitless. Here we can use regression algorithms to solve this problem. In regression we do not have to label data into different classes. Instead, we just have to predict a final outcome. Another example for regression is predicting the price of stock over a period of time. In supervised learning, the machine already knows the output of the algorithm before it starts working on it. In here machine is labeled with both output and input.
Unsupervised Learning
In here the data is not labeled and have no guidance. The machine has to figure out the data set given and has to find hidden patterns in order to make predictions about the output. In unsupervised learning the machine is trained on unlabeled data without any guidance. Unsupervised Learning can be used to solve association problems and clustering problems. Association problems basically involve discovering patterns in data finding co-occurrences. Clustering is used for cases that involve targeted marketing. Example for clustering is, if a data of list of customers given, the machine should be able to cluster these customers based on their similarity. In unsupervised learning, the machine is only given the input data.
Reinforcement Learning
In reinforcement learning, an agent interacts with its environment by producing actions and discovers errors or reward. The main difference in reinforcement learning than others is that the input itself depends on the actions taken. For example assume a situation where a robot does not know anything about the surrounding it is in. After it performs certain actions it finds out more about the surrounding. In this case, the robot is known as the agent and its surrounding is the environment. There is no predefined data for reinforcement learning. The input is only dependent by the actions taken by the agent. These actions are recorded in the form of matrices so that it can serve as a memory to the agent.
Real World Machine Learning Usages
- Predictions -Traffic Predictions, Weather Predictions
- Social Media Services -Face Recognition
- Product Recommendations
- Search Engine Result Refining
- Email Spam and Malware Filtering
- Virtual Personal Assistants-Amazon Echo, Google Home
