Machine Learning: A Beginner’s Guide to What is it, How it Works, and What to Expect from it.

Prerequisite: Basic knowledge of computer science terminologies like algorithms, programs, etc.

Have you ever spent a day watching ‘The Office’ clips on YouTube and the very next day you have the same show’s clips recommended to you? As simple as it is to think that the program just understands your likes and dislikes, behind the scenes, it’s actually quite a complex algorithm. But, hey, you have come to the right place! This concrete article will get you to understand the idea behind it all — Machine Learning, and help you not feel left out when the buzzword pops up in any conversation.

Machine Learning — the two words that have taken over the tech industry by storm in recent years. Well, what is it?

To understand Machine learning, it is necessary first to understand why computers came into the world in the first place. You see, humans are at the very least — humans i.e. we lack absoluteness. As the desire for faster, more accurate, and consistent calculations-based tasks increased computers quickly became the driving force of the industrial world. It became irrational to deploy a human workforce to solve problems that can be solved by a computer much more efficiently.

Let’s say a company wanted to hire the most experienced candidate for the role of Manager. Then, they could simply program the computer to compare the years of experience of all the candidates and find the most experienced one. Something that could take quite a few man-hours to skim through will be done in a matter of seconds. This started saving a lot of time and most importantly, money, for businesses.

But let us take this a little further, what if we were required to write a program that would tell us if the given item is a picture of a dog? Certainly, we can instruct the computer to check if it has four legs or, if it barks. But then we would have to teach the computer what bark means. This goes on and on and becomes much more complicated to program.

Earlier, we required a human to make such decisions on behalf of computers, but now this is where Machine Learning casts its spells.

Stanford University defines Machine learning as — the science of getting computers to act without being explicitly programmed.

Simply, Machine learning is training a computer on the best available data to perform a certain action(like training an autonomous car to identify obstacles by providing images of obstacles and non-obstacles as input data).

But how does it actually work?

The computer or machine takes in data and trains up a model to work through a problem. The computer may be supervised or unsupervised during the process. During supervised learning, both input data and the ideal outputs are provided ( input: an image of a fallen tree, Output: Obstacle). In unsupervised learning, only input data is provided (to find hidden patterns within data, not suggested to use for autonomous car case). Either way, it keeps learning and updating the model to work on as new data become available. To paraphrase it in simple steps :

• The machine keeps feeding on data and runs experiments on them
• It finds a pattern within that data through these experiments
• When new data are given, it can predict the outcomes based on that pattern.

These steps are key for any Machine Learning process.

Fig. Machine Learning Illustration for an Autonomous Vehicle System simplified(created using Figma)

Data Science and Machine Learning

Notice how it all revolves around data? This is why you keep hearing about data science alongside Machine Learning. Data Science can be taken as a much broader field that applies scientific techniques to extract insights from data. Machine Learning, on the other hand, is a large subset of AI at present. These two fields will always be interlinked and go hand in hand in the development of Artificial Intelligence.

The next big thing?

Although the use of machine learning technology is still in its early stages, it has already been integrated into our daily lives through various applications such as virtual assistants like Siri and speech-to-text converters like Google’s. While we are currently only seeing the beginning of what this technology has to offer, it is likely that we will continue to see an increase in the use of machine learning in various aspects of our daily lives in the future. It is just the tip of the iceberg in terms of the potential applications and advancements that we can expect to see in the coming years.

So there you go ! That’s all you need to know to get started in the field of Machine Learning. The next time you come up to something you like on YouTube, you know how the strings are being pulled behind the scenes.