0-1. Understanding The Difference Between AI, Machine Learning And Deep Learning!
Welcome to AI-SCHOLAR's new content for business people, "Learning AI Systematically"! As a start to learning, this section describes the "AI I'll explain what the word "I" means and how you should understand it! This section will help you learn the difference between AI, machine learning, and deep learning.
1. the AI isn't magic!
The term "AI" has exploded in popularity in recent years.
AI It's good that the knowledge about it's also true that the content is littered with unclear statements and sometimes errors. In the business world, we hear about the success of AI, but we don't know what it can actually do, and the words are just getting in front of us.
However, there is actually no clear definition of AI in academia.
He was a professor at Stanford University in the United States. Computer scientist Arthur Samuel. field "a field of study that gives computers the ability to learn without being explicitly programmed to do so. Perhaps the most famous explanation is.
Arthur Samuel [Image source: Stanford InfoLab]
If we take this description to heart, we might ask, "Can we magically manipulate and create tools without the need for programming or formulas?" Some people may think that this is a good idea. However. This definition is the ideal form of AI proposed long ago, and current AI technology has not yet reached this level.
And although AI is a field of research, it is now often used in the sense of a function. Many companies are now using this calculation method in their businesses to achieve success.
If I were to venture to explain AI at the current level of practical application, I would say 'The ability of a program to analyze features from large amounts of data and predict and classify them with a high degree of empirical accuracy.' I think this is the right way to describe it.
For example, the AI that defeated the world chess champion can predict and select the move with the highest probability of gaining an advantage by learning a large number of strategies and experimenting with them. For example, an AI that automatically recognizes faces would need to be able to predict and select a move with the highest probability of advantage by learning a large number of strategies. The images of various people's faces are classified based on their tendency to learn what a "human face" looks like by learning We judge by doing.
In other words, while AI technology can be "used" for many things There's not much that can be done by AI itself alone, and it's still being used by humans This is the current state of affairs. As you read through this content, you will surely learn more about the functions of AI and its limitations.
There is a big gap between the image of AI as it is generally perceived and the image of AI as it is actually being used by companies. Correctly understanding this gap is the first step to getting into the AI business. In fact, AI has become a very hot word for researchers around the world. Now is the time to learn the basics so that you don't get left behind in the fast-paced AI business that is sure to grow at an alarming rate in the future!
2. machine learning? Deep learning? First, let's know the difference!
Now, we've found it difficult to explain the definition of AI in one word.
By the way, there's more than just AI. Machine learning and deep learning I've heard a lot of words such as "I don't know what to do". These are words that relate very strongly to AI, but first We'll keep the AI and related terms down so that you can understand the distinction between them well. Let's be.
AI, Machine Learning, Deep Learning (Deep Learning ) The relationship between the two can be illustrated as follows. Artificial Intelligence (AI) is the one that contains the most various meanings This is why the term "AI" is so broad in scope and very difficult to understand if it is used crudely.
The relationship between AI, machine learning and deep learning
Now, here's a quick rundown of Let's talk about what machine learning, deep learning, is all about.
◾ Machine Learning
Machine learning may be easier to understand if we define AI as "a system that automatically performs calculations based on the ability of a program to analyze features from large amounts of data and make predictions and classifications with high accuracy."
Since machine learning refers to the computational system within AI, it is often treated in much the same way. However, it is important to understand the differences between the two.
For example, we can say that image recognition capabilities, chess AI, etc. were also created using machine learning systems.
◾ Deep Learning
Machine learning is a method of finding regularities and relationships in large amounts of data to make decisions and predictions. In order to do this, like "color" and "shape, It is necessary for a human to specify the characteristics (parameters) to be focused on.
Deep learning, or deep learning, is the A field within machine learning that adds a new mechanism to machine learning It may be easier to understand if we say that this refers to
Modeled after the human brain's neural circuits Multilayered Algorithm "Deep Neural Network The AI (artificial intelligence) itself considers and decides on the settings and combinations of features using a variety of methods. In machine learning, you had to direct your attention to things like "color" and "shape," but deep learning In the case of (deep) learning, it can figure out for itself what values should feature and make the best classification or prediction without having to tell it what to do. I won't go into detail here, but it's a very interesting principle.
And, most importantly, so as not to mislead you " deep learning that the data is also necessary (without it, nothing can be done) and that there are operations that humans can add to it." must be aware of.
Currently, it is not necessary to understand in detail how "machine learning" and "deep learning" work. In this section, we will discuss their inclusion relationship and how AI Programs to predict and classify features based on data Let's keep in mind that it is
When you're done reading, press the button.
Let's track our learning progress!