Surely when reading this you do not think about how it is that your brain is capable of transforming what you are seeing into useful information, something that, on the other hand, has long been investigated by human beings. Philosophers, scientists and computer scientists, all of them have been working in part to find out how the human brain works, to which they have been able to respond with incredible advances with which today, thanks to technology, we can advance even faster in this field and we are even in an era in which we can replicate complex human behaviors through technological devices, which is the topic that we will focus on today in this article.
In recent decades, advances in the exploration of human behavior have had amazing repercussions, and new terms and disciplines have been developed that allow for greater exploration and investigation.
Today we will talk about machine learning, one of the most important fields of technology when it comes to studying human behavior and replicating them through technology.
But before this we need to understand more basic concepts, which brings us to our first stop:
Artificial intelligence
Probably if you had heard this term not many years ago the first thing that would have come to your mind probably would have been some science fiction movie in which machines dominate the world and destroy us all, but it is just another stereotypes that hollywood has instilled in us through the years.
Currently and more now when we are surrounded by so much technology and information, artificial intelligence is more than fiction and frankly without artificial intelligence, the panorama would be bleak.
But and to all these …
What is artificial intelligence
According to Wikipedia, artificial intelligence is:
“Artificial intelligence is intelligence carried out by machines.”
But this definition falls short when talking about a field as extensive as artificial intelligence, so:
“Artificial intelligence is the sub-discipline of the field of computing, which seeks to create machines that can imitate intelligent behavior”
But what happens with this definition is that saying “imitate intelligent behavior” may not be very appropriate since in some cases replicating intelligent behaviors is not very complicated, for example there are some robberies that can walk and this in turn is behavior. smart but this robot was programmed to do this move, it didn’t necessarily learn to walk on its own, so the definition of artificial intelligence is a bit fuzzy when you consider all the possible cases where technology can fit this definition , which is why within artificial intelligence there are too many subfields or categories, but today we are specifically interested in one:
Machine Learning
What is machine learning
“Machine learning is a method of data analysis that automates analytical model building. It is a branch of artificial intelligence based on the idea that systems can learn from data, identify patterns and make decisions with minimal human intervention.”
According to the previous definition we can deduce that machine learning is one of the branches of artificial intelligence, machine learning is one of the most important subtopics when it comes to talking about simulation of human behaviors because it allows the machine or the software learn for yourself without the need to do it explicitly in the source code.
Machine learning is the field that ensures that machines can draw conclusions based on experience, through a number of incoming data or experiences. Example
As a programmer, I can tell a machine that the photos tagged with the name “dog” will be classified as a dog, bearing in mind that if I pass a photo without a tag, the program will probably generate an incorrect answer, but thanks to machine learning I can provide the machine with photos of dogs with the label of “dog” so that when I pass an unlabeled image of a dog, he can deduce that what happens to him as data is a photo of a dog and thus the machine can automatically sort.
It is a technology that allows a series of operations to be automated in order to reduce the need for human intervention. This can be a great advantage in controlling a huge amount of information in a much more effective way.
Why is important
Machine learning is important in many cases, even more so in the times we are in since machine learning allows us to automate too many things that as human beings would take us a long time or we simply could not do, machine learning has multiple uses, in medicine for example, recognizing patterns in data that at first glance seem confusing, such as detecting some types of cancer cells quickly and accurately, as well as in the automotive world, such as smart cars that drive themselves, in the field statistics by reviewing giant databases and tying loose ends between them, as well as a host of possible uses that can be given to this technology.
How does it works
Normally in classic programs or machines the people who develop them create a series of algorithms which allow them to do a specific task which is already predefined when programming.
On the other hand, the algorithms used in Machine Learning perform a large part of these actions on their own. They get their own calculations based on the data that is collected in the system, and the more data they get, the better and more accurate the resulting actions will be allowing the program to self-program as needed, obviously with some limitations.
Algorithms that allow machines to classify data in order to automatically convert it into useful information are called classification algorithms.
Machine learning types
Supervised learning
This type of learning is based on what is known as training information. You train the system by providing it with a certain amount of data, defining it in detail with labels. Example:
You can have a database of images of apples and pears but they are not classified, what you do with supervised learning is to pass to the machine some photos of pears and apples with their respective labels so that the program can do it automatically after taking Take into account those images with which he was trained.
Unsupervised learning
This type of learning, unlike supervised learning, is in charge of collecting information that is automatically implicit in a group of data, for example:
There are 100,000 images of cells and with this model it is intended that the machine can deduce for itself which type of cell have conditions only by means of a large amount of data.
Reinforcement learning
In this type of learning, the machine or program learns from a system of rewards and punishments. As an example:
you can observe the behavior of an autonomous car. When the vehicle makes an erroneous decision, it is penalized, within a system of registration of values. Through this system of rewards and punishments, the vehicle develops a more effective way of performing its tasks
![example4]()
To conclude, I would like to say that machine learning is one of the most important branches of artificial intelligence since it allows us to understand more the behavior of the human being, in addition to helping us in the automation of infinity of things that years ago were impossible, and although I think its progress to date is amazing, we must bear in mind that it is a new discipline in quotes and that it still has a long way to go, as there are still too many things that are not within our reach today and that probably in a few years we can do and we will be able to have even more present that field of technology in our day to day.
I hope this information has been to the liking of all the viewers, and RTFM :)