AI, ML & DL
What's the difference between Artificial Intelligence, Machine-Learning and Deep-Learning?
Artificial Intelligence (AI) often evokes science-fiction movies like Her, Ex-Machina or Matrix, but the technology is no fiction anymore, it is real, and its usability is within reach of anyone who owns a smartphone or computer. Big companies like Google, Apple, Facebook and a lot more have hopped on the train of AI and are using its perks
An overview of the themes:
in their apps or devices, proving that the use of cognitive computing improves many simple routine tasks and efficiency.
But even though AI has become a hot topic in medias, the extent of its development is still a little vague for some. Fortunately, you can narrow it down to two very important concepts: machine-learning and deep-learning. What do they mean and how do they work together?
What is to be understood under the term of AI?
The simplest explanation would be to say that it’s the ability for machines, to show human-like intelligence. As if a computer had a brain of its own, learning, planning and recognizing objects, languages and images.
AI can be separated in three different types: Artificial narrow intelligence (ANI), artificial general intelligence (AGI) and artificial super intelligence (ASI). Whereas ANI is maturing and a work in progress, AGI will probably only be introduced in a couple of decades and ASI will take even longer. This is why we’ll only concentrate on ANI for this article.
Narrow AI, also called “Weak AI” is what can be used in our daily life (Siri, Google Assistant). Contrary to what you can see in some movies, this AI shows no human traits, it has no conscience, no emotions and no opinion. This type of AI can overperform the human brain but only on single tasks. It is an algorithm programmed to excel in a predefined field (playing chess, searching something on google, recognizing images etc..).
ANI can’t go out of this field, it can only operate inside of it. Even though, it is called “weak”, its response is always pretty accurate and precise.
It is for now, the only type of AI that is in use, it is being developed, it is maturing but it’s still miles away from AGI (artificial general intelligence).
What is to be understood under the term of Machine-Learning?
To put it in simple words machine-learning (ML) is a category of algorithm that interprets data, learns from the data, and applies said learned data to make informed and accurate decisions. Machine-learning is a progressive learning process, it performs a function and the more data and input it gets, the smarter and better the function will become.
As an example, we could use any App that has contact with its users. Spotify for an example will give you recommendations based on your music tastes. You can upvote or downvote these musical recommendations and from that input, the algorithm gets “smarter” and so the recommendations a lot more accurate.
What is to be understood under the term of Deep-learning?
To get machines to learn you need deep-learning (DL). Deep-learning is a subset of machine learning and even though it is intricately bound to it, their functions and roles are different. Deep-learning is inspired by biological neural networks based on the human brain and by its use, it makes standard ML models more capable and independent. Where a machine learns, deep-learning helps it think and draw conclusions.
How does it work? Machine learning models do get progressively better, but they need help and guidance. Normally an engineer or programmer would check the output of the algorithm and fix the issues, but with a deep-learning model the algorithm can determine on its own if a prediction is accurate or not, it learns from its own method of computing, its own “brain”.
How do they work together?
Even though the three terms are different and engage in different types of work, we could say that AI englobes all three of them.
AI describes the concept of human-like intelligence in machines while ML and DL help it to learn, to conceptualize, to analyze and recognize. One doesn’t really go without the other. Where AI aims to increase the chance of success, ML aims to increase accuracy.