Artificial Intelligence and Machine Learning in layman term

4 min read4 hours ago

Artificial Intelligence

In this booming era of Artificial Intelligence (AI), everything is integrated with AI. Every new products on the market, be it a smartphone or a washing machine will try to integrate this golden technology in order to keep up with the AI-this-and-that trend of current market. Ask any Computer Science students on what their dream project is, and their answer will be _____ with AI. This phenomenon of wide adoption of AI on different industries is compared by Andrew Ng, an AI expert to Thomas Edisons’ harnessing of electricity which revolutionize dozens of industries at that time. When talking about AI, a term; Machine Learning (ML) always follows around. The term AI and ML are always used interchangeably, up to a point where one might think that both are the same thing — which they aren’t — and a quick google search will eventually tell us that ML is a subset of AI. Hence the question is, what is Machine Learning (ML)?

In order to understand what machine learning is, we must first understand what AI is. IBM defines AI as:

Technology that enables computers and machines to simulate human learning, comprehension, problem solving, decision making, creativity and autonomy.

In a simpler term, just like artificial sweetener where the substance mimics the sweetness and taste of sugar; AI is a technology where a computer or machine mimics human intelligence. Human intelligence itself is defined by Brittanica as:

Mental quality that consists of the abilities to learn from experience, adapt to new situations, understand and handle abstract concepts, and use knowledge to manipulate one’s environment.

Basically, human intelligence is the ability to learn any kind of knowledge, and apply it in an appropriate manner.

Machine Learning

Now come the part where we’ll be introduced to Machine Learning. Remember that the goal of AI is to mimic human intelligence; in mimicking the ability-to-learn-any-kind-of-knowledge part of human intelligence, the domain is known as Machine Learning (ML). ML is defined by IBM as:

A branch of AI and computer science that focuses on the using data and algorithms to enable AI to imitate the way that humans learn, gradually improving its accuracy.

Keep in mind that the whole idea of AI is to mimic human intelligence; same goes to ML, it try to mimic how human learn. There are a few types of machine learning algorithm, but 2 of the most well known are; Supervised and Unsupervised learning. And just like human, when we want to learn, we must have a goal on what we want to learn. To further understand this, let’s look at an example; Consider a setting: A teacher want to teach a student on how to differentiate between a triangle and a rectangle:

Supervised learning

In a supervised learning the teacher will explicitly tell the student which shape is triangle and which shape is the rectangle by labeling the shape (triangle and rectangle). Then, the teacher will give him some samples of combination of both triangles and rectangles shape and ask him to label which is which. If he labels it wrong, the teacher will correct him. As time goes by, he will become an expert in differentiating between a triangle and a rectangle.

2. Unsupervised learning

In an unsupervised learning, the teacher will give the student the triangle and rectangle, and let him analyze both of the shape by himself (note that the student don’t really know the end goal is to differentiate between a triangle and a rectangle). Eventually, the student will find out that there are some distinctions between the two shapes; the number of sides, the shape of it, the angles, etc. Based on those, the student then be able to deduce that in this sets of given data (the 2 shapes), there are 2 categories which implies to rectangle and triangle.

In this example, the teacher is the researcher (human), and the student is the machine learning algorithm. The output of this process is what we call as an AI model. Integrate this model with a beautiful user interface and voila! we have an AI-powered software ready to be used!. This is essentially how any AI-based-products like chatGPT are operated; There is a trained AI model on the background that analyzes the input and return an appropriate output.

In conclusion, Artificial Intelligence and Machine Learning are two closely related terms, but aren’t referring to the same thing. Artificial Intelligence is a broad term referring to the ability of machine to imitate human’s intelligence; Machine Learning is a part of Artificial Intelligence that focuses on imitating the learning-ability of human — by training ML Algorithms to produce AI models. Saying that AI and ML is the same thing is not entirely wrong, but hey! let’s give this field a respect by using the correct term! <insert nerdy face emoji + finger pointing up>.

Note that this is a simplified explanation of AI and ML. Check out the references list down below for a more proper and detailed explanation.

Artificial Intelligence and Machine Learning in layman term

Artificial Intelligence

Machine Learning

References:

Written by Wahba Kamaluddin