Introduction To Machine Learning - Part One

Machine learning is a field or a part of computer science that gives the robot or computers the ability to learn without being explicitly programmed. Machine learning is employed in a range of several computing tasks where designing and programming explicit algorithms with very good performance are difficult or infeasible.

Machine learning is closely related to computational statistics which is very famous or focused on prediction making through the use of computers. It consists of strong ties to mathematical optimization, which delivers methods, theory, and application domains to the relevant fields.

Machine learning is a method which is used to devise complex models and algorithms that lend themselves to the prediction, also called predictive analytics.

Supervised Learning

In supervised learning, we have a data set and we already know what our correct output should look like. Also, we have the idea that there is a relationship between input and the output.

Supervised learning problems are categorized into “regression” and “classification” problems.

  • Regression Problem
    In a regression problem, we try to predict results within a continuous output. In short, we try to map input variables to some continuous function. For example, we have a picture of people and we have to predict their age on the basis of the given picture.

  • Classification Problem
    In a classification problem, instead of trying to predict results in a discrete output, we try to map input variables into discrete categories.

    Here, we know we have a patient suffering from a tumor. We have to predict whether the tumor is malignant or benign.

Unsupervised Learning

In this type of machine learning, it allows us to approach problems with little or no idea; that means we don’t know what our results should look like. We can derive the structure from data where we don’t necessarily know the effect of the variables. The other way is that we can derive this structure by clustering the data based on relationships among the variables in the data.

With this type of unsupervised learning, there is no feedback based on the prediction results.

Machine learning

Reinforcement Learning

In this learning, the learning is more concentrated on how an agent ought to take actions in an environment so that it can maximize some notion of long-term reward. It is totally different from supervised learning problem in which we never provide correct input and output, not sub-optimal actions explicitly corrected.