A brief about Machine Learning
The term machine learning was popularized by Arthur Samuel in 1959, who was a pioneer in computer gaming and artificial intelligence. He defined ML as “field of study that gives computers the ability to learn without being explicitly programmed”. Simply put, ML is the ability of a computer to learn from data to make future predictions.
How do machines learn?
Things which come very naturally to humans can be very difficult for machines. You only need to teach a task a couple of times before a human can do it on their own. But, in the case of machines, we need to collect a lot of data in order to teach machines to perform a specific task. Machine learning uses data and relies upon different algorithms to find patterns in the data and “learn.”
Some examples of machine learning in daily life:
- Amazon and Flipkart recommending products based on your past purchases
- Facebook and Instagram recommending posts you might be interested in
- Swiggy and Zomato showing you the ETA for food delivery.
Now that we have a basic understanding of what machine learning is, let’s look at the types of Machine Learning.
Types of Machine Learning
Supervised Learning
We use a labelled dataset — which means we have both the input data as well as the desired output. Model learns from past results and predicts the future.
Regression: Regression is used when the output is a continuous variable. Examples where we can use regression:
- Predict the price of a house based on its features
- Predict the estimated time of arrival
Classification: Classification is used when the output variable is discrete or it is a category. Examples:
- Image classification
- Classifying emails as spam or non spam
- Approving the loan of customer based on credit history
Unsupervised Learning
In unsupervised learning, the outcomes are unknown. The model groups the data into different clusters wherein, the members within each cluster are similar to each other.
Examples:
- Grouping customers with similar behavior based on customer characteristics and past buying records
- Document classification : Identifying the main theme of documents and classifying them.
- Classifying network traffic : It is beneficial to know where the traffic is coming from
Reinforcement Learning
In reinforcement learning machine learning model is trained to make a sequence of decisions. The agent learns to achieve a goal in an uncertain environment. The computer employs trial and error to come up with a solution to the problem. To get the machine to do what the programmer wants, the agent gets either rewards or penalties for the actions it performs. Its goal is to maximize the total reward.
Applications of reinforcement learning:
- Autonomous vehicles, self driving cars etc.
- Traffic Light Control
- Gaming
Resources:
- www.almabetter.com — course materials
- https://deepsense.ai/
