Table of Contents
In todays world machine learning is one of the fastest growing fields in technology industry. According to google trends machine learning related keywords are highly searched globally in the past few years. By 2030, the global AI and Machine learning market is expected to reach $1.6 trillion. Indeed, also ranked machine learning engineers as one of the top 10 jobs in the world with average salary of $ 140,000 annually in United States.
So, the question now is what is machine learning and why is it so important in today’s world? I have read a lot of articles and watched many videos to get the idea of machine learning. So, to make this simple for you I will divide this topic into 5 main subtopics so that you will be able to understand the concepts and uses of what machine learning is actually about.
Before diving into a roadmap and start learning machine learning I highly recommend to explore the following topics and a get an idea of what you are about to learn. All these subtopics of machine learning covers only the basic and there are no deep concepts so that even a kid could understand how machine learning works after exploring the following articles.
4. Machine Learning Tools and Mathematics
5. Machine Learning Engineer Roadmap For 2025
Now let’s dive into the world of machine learning, Happy learning all!
What is machine learning?
Machine learning is turning data into numbers and finding number patterns in those numbers. Finding patterns is done by computer using maths.
Machine learning is the field of studying that gives computer the ability to learn without been explicitly programmed – Arthur Samuel
Difference between Traditional programming vs machine learning
Let’s see a real-life example of creating a chicken soup to understand the main difference of traditional programing and Machine Learning.
Traditional Programming
Machine Learning
As you can see in traditionally programming we have ideal inputs and we process that input data to get the meaningful output. But in machine learning we usually have inputs and the ideal output and the ML algorithm figure out the process (Turn inputs into numbers and ML algorithm figure out the patterns in that numbers to match the relevant output)
However traditional programming can be done without machine learning but machine learning cannot be done without traditional programming.
Why use machine learning
If you can launch a product based on simple rules that doesn’t need machine learning you probably do that (rule 1 of google machine learning). But if you can’t think of all the rules you probably go for machine learning.
Where should machine learning will be needed??
- Problems with long lists of rules
- Continually changing environment
- Discovering insights within large collection of data
Let’s take an example of self-driving car to understand the above scenarios better
Usually a self-driving car turns the image ahead to numbers. As examples, it identifies the road lines as 101, another car as 111. And then it decides what to do when these numbers appear. If it sees another car it avoids turning into the side of that car, if it sees road line it decides not to cross them like that. So, if you were to design this self-driving car system and if you were to code this all these rules from yourself that would take a very long time. This is where machine learning comes to play – Problems with long list of rules.
And if you hand coded this self-driving car to work for city road, when it enters a tunnel the system will not be working. This is where machine learning comes to use again – Continually changing environment.
So, you get these images from the camera of self-driving car (inputs) and decide what actions needed to be taken for from the car itself (outputs). After getting those inputs and outputs we need to create an algorithm and deploy it back to the car system. This is where machine learning will be good at again – Discovering insights within large collection of data.
Conclusion
Machine learning is a powerful technology that is reshaping industries and solving complex problems. By learning about its basics what it is, the difference between traditional programming and machine learning, why machine learning is needed—you’ve taken the first step in understanding this fascinating field.
This is just the beginning of your journey into machine learning. In the next article, we’ll explore the different categories of machine learning problems, giving you a deeper understanding of how Machine Learning works in real-world scenarios. See you there!
