25 Oct What is Artificial Intelligence?
This article gives you a non-tech introduction to AI and what it means in layman terms. You will get introduced to the most used types of AI.
Everyone has heard of artificial intelligence, but it is probably difficult for many to explain what the domain is all about.
The definition I like best is to define artificial intelligence as not human intelligence—meaning, intelligence that originates from machines, as opposed to natural intelligence that originates from humans.
The Concept of AI
The concept of artificial intelligence is often used to describe machines that perform cognitive functions such as problem-solving or learning. What this means in practice, I will return with a little later. First, the definition needs to be refined a little bit more.
When reading articles on artificial intelligence, you often encounter different words and concepts used to describe the field. The three that you will most frequently face are:
- Machine Learning
- Deep Learning
The short answer to the difference between them is that AI is a collective term for all Artificial Intelligence. Machine Learning is a (popular) branch of artificial intelligence. Deep Learning is another branch of the Machine Learning universe. So, mathematically, it is a combination of real subsets, as shown in the figure below.
The three concepts
The significance of these three key concepts is:
- AI: Is the broadest term for artificial intelligence. It means that it covers everything from conventional AI methods to advanced Machine Learning and Deep Learning techniques.
- Machine Learning: Is a subgroup of AI that covers algorithms that learn by being trained with data. Countless Machine Learning techniques exist. If you come across words like linear regression, decision trees, then they are all Machine Learning techniques.
- Deep Learning: Are algorithms that if they get enough data can become self-learning. A Deep Learning algorithm works in a way that is inspired by the human brain. The way Deep learning is designed is architecturally is called Neural Networks.
The area that is growing the most by far is Deep Learning. Many of the new revolutionary AI inventions that we almost daily hear of are made as Deep Learning solutions. So, understanding what exactly Deep Learning can do for you is significant.