Can Machines learn – The technology of Machine Learning

by Kavitha — December 16, 2021

All of us learn a lot every day…we observe and learn, we are taught by our teachers and parents and thus too we learn… probably we learn all throughout our lives. This is thanks to the wonderful brain that mankind is blessed with. If I were to tell you that we are in an era where it is not just the ‘brainy’ humans who learn, but that they have managed to devise methods to even make machines learn, isn’t that awesome?

Well, it is not a sci-fi fantasy but it is true. As true as the Alexa you turn to, as true as the auto predict tools on your devices. Yes – we are talking about something called Machine Learning.

Unlike humans, baby machines do not have to go to a school to learn what they ought to learn. But before we get into that let us first understand what are machines?

Machines are devices that can make our work easier. Did you know the cell phone and computer you use is a machine? So are cars and other devices in and around your house. Now let us get back to what makes the human mind so capable of learning. Neurons or nerve cells are the primary components of the brain which aids in learning and memory. They form an intricate network among themselves which forms the basis of learning and this neural network is what is exploited in making machines learn too. In Machine learning, interconnected processors of external inputs act as these neurons which secure information to train the machine.

PC: Rotimatic

Algorithms are a packet of instructions arranged in a structured manner such that one needs to take the input and process it into the loop and receive the output at the other end. Let us understand this in the context of a modern automated roti maker. One takes wheat flour, salt, water, and oil and adds it to the input funnel, feeds in data on how many rotis need to be made, and start the machine… and Voila, hot fluffy rotis are delivered at the outlet. For those uninitiated in Indian cuisine rotis are flatbreads. Well what we observe here is that the machine or the roti maker has a series of instructions programmed inside it to mix, knead the ingredients, divide it into smaller portions, flatten it out and then cook it before producing the rotis so effortlessly. This series of information that the roti maker is fed with while manufacturing is much like what an algorithm is. A set of instructions that can be blindly followed to receive the desired output.

Neural networks and Algorithms form the basis of Machine learning. Algorithms are trained based on the information in the neural network to make the right decision instead of blindly following what is pre-programmed into it.

Now that we know what is machine learning, let us see what are the possible applications of this revolutionary branch of technology. Some examples you may have come across include:

Automatic Recommendations– You might have received suggestions of videos, songs, webpages that you may have searched for earlier, or even noticed this in the recommended products section of Amazon, flipkart or other e-commerce sites based on what you had been searching for earlier. The machine was learning your preferences as you worked on it and now takes the liberty to recommend to you, products or services you may like.
Predictions: Prediction of outcomes for a variety of tasks can be done using machine learning using data obtained from past events, studying its pattern and then predicting the outcome (I wonder whether Paul the Octopus was the Father of Machine learning for prediction 😛 ) Prices of stocks, movement of players in sports, etc can be predicted with a fair bit of accuracy using this technology
Smart cars: which can sense and assess the driving conditions with the help of sensors and learn to take decisions on when to operate the brakes, accelerator etc.