Introduction to Machine Learning and Artificial Intelligence
What is Machine Learning ?
To simply put it , Machine Learning is basically making a computer learn various intuitions from large amounts of data and using it to perform specific tasks ,without the computer being explicitly programmed to do so.
It might sound like magic. but , It’s nothing but matrix multiplication and basic differentiation
Types of Learning
To understand better , Let’s see the common ways with which a computer can learn. So, learning is classified into 3 parts:
1. Supervised Learning
2.Unsupervised Learning
3.Reinforcement Learning (Advanced)
Supervised Learning
Supervised learning is a type of learning where Each input data in the data-set is mapped with a corresponding output label in the same data-set.
This definition is right. But, it is way too abstract. So, let’s see some examples of problems that come under supervised learning for better understanding. Supervised learning problems are further classified into Regression and Classification problems
Example 1: Given the experience of a software engineer as an input, predicting the salary of the software engineer — (Regression Problem)
Example 2: Given the size of a tumor , classify whether it is benign or malignant (benign and malignant are 2 types of tumors where malignant is dangerous and more risky than benign) — (Classification Problem)
Example 3: Given the features of a house as an input, such as area, number of bedrooms, etc., Predict the price of the house — (Regression Problem)
Example 4: Given the details of a customer , Predict whether a loan should be given to the customer or not — (Classification Problems
So , these are the 2 types of supervised learning problem
Unsupervised Learning
Unsupervised Learning is a type of Learning where the computer learns information about the data only from the input data.
In other words, There is no output label mapped with the input data.
Again , It is is kind of difficult to interpret Unsupervised Learning just from this definition.
So, let’s dive into some examples and look at problems that help us better understand Unsupervised learning
As far as I know, unsupervised learning is used for a particular type of problem called clustering.
Just by looking at this data , It is easy for us , humans to understand that the data in this plot can be split into 3 clusters. but a machine understand nothing but numbers. so , inputting large amounts of data and making the machine cluster the data by itself is called the Clustering in Machine Learning.
One thing to notice here is that we are not going to specify the cluster to which a particular data belongs to, like we did in Supervised learning problem.i.e, There is no target label
Example of Clustering : Can be used to segment different types of product in Marketing
Reinforcement Learning
It is an advanced type of learning compared to the other types of learning in Machine learning and I would consider this as the true form of Artificial Intelligence.
It is Difficult to Explain Reinforcement Learning in words.
But, The basic concept is that there is an Agent which is present in an Environment and the agent is rewarded based on the action it takes in an environment. It’s exactly like a video game and the goal of the agent is to maximize it’s reward
It would be better to understand Reinforcement learning by watching some videos. So, I will post some links down below:
Examples of Reinforcement Learning:
Example 1: AI learns to play snake
Example 2: AI learns to play Hide and Seek
Conclusion
This is a very basic introduction to Machine learning. But, there are still a lot of algorithms for these problems that cannot be covered in a single blog post.
That’s it. Thanks for reading.
