Machine Learning : Introduction

Machine learning is a crucial branch of the field of Artificial Intelligence (AI). If AI is the dream of creating human-like intelligence, then machine learning is one of the key tools to make it happen—by giving computers the ability to learn independently, without having to be explicitly programmed for every single task.

The core of machine learning is simple: learning from data. In the world of ML, data is everything its fuel, its teacher, and its foundation. The more high-quality data we provide, the "smarter" and more accurate the resulting ML model becomes. This process is similar to how humans learn from experience, only it's done by computer algorithms analyzing patterns in very large datasets.

There are a few main approaches in machine learning, each with its own unique method. 

• Supervised learning works like a student learning with a teacher and an answer key; we give the algorithm labeled data (for example, pictures with captions like "this is a cat" or "this is a dog") so it can learn to identify patterns. 
• Unsupervised learning is more like an explorer trying to understand a new city without a map; the algorithm is given unlabeled data and its task is to find hidden structures or groupings within it.
• Reinforcement learning resembles animal training or a gamer's learning process: the algorithm (an agent) learns through trial and error and feedback in the form of rewards or penalties from its environment to find the best strategy. We will discuss each of these methods in more detail in the next material.

The application of machine learning in the real world is already very broad and something we often encounter in daily life. Product recommendations on e-commerce platforms like Tokopedia or Shopee, facial recognition features to unlock phones, virtual assistants like Google Assistant, traffic prediction on Google Maps, to spam filters in our email all of these are powered by machine learning. Even in more complex fields like healthcare (to help analyze radiology), finance (for credit scoring), and agriculture (to predict harvest yields), ML is starting to play a major role in increasing efficiency and accuracy.