What is Machine Learning?
The first programs based on AI, such as Deep Blue, were based on rules and programmed by a person. Machine Learning in a branch of artificial intelligence that began to gain importance in the 80’s. It’s a form of AI that no longer depends on rules and a programmer, instead the computer can establish its own rules and learn by itself.
When Google’s DeepMind managed to beat the Go world champion, it did so my applying machine learning techniques and training itself with a large database that gathered information from the games of expert players. Therefore, it’s a great example of the application of ML.
ML systems work on large volumes of data, they identify patterns in its behavior and are capable of predicting future behavior based on these patterns. In this way, they are capable of identifying a person by their face, understanding a conversation, detecting an object within an image, translating text and much more. It is the most powerful AI tool for businesses. As such, large technology companies such as Amazon, Baidu, Google, IBM, Microsoft and others offer their own “ML For Business” platforms”.
How do machines learn?
Machine learning is the result of algorithms. An algorithm is nothing more than a series of ordered steps that are undertaking to complete a task. The aim of ML is to create a model that allows us to solve a given task. Then, this model is trained with large quantities of data. The model learns this data and is capable of making predictions.
Depending on the task that it wants to carry out, certain algorithms will be more appropriate than others. The models that we obtain depend on the type of algorithm that we choose. We can work with geometric, probabilistic or logical models. For example, one of the logical models most well-known is the one based on the decision tree algorithm.