Essay on automatic learning
INTRODUCTION
Today technologies have been put as a very important and special tool for humans, so thanks to them we can efficiently perform our work work, so much that now the devices have the ability to learn, reason, understand andform an idea of reality, this has generated that users be dependent on these technologies.
Thanks to these tools they allow the human being to extend and enhance their skills, with a set of knowledge, processes and actions.
Automatic learning is very helpful in so many companies and organizations since computers already know that products like customers thanks to the valuations they give to each product.
In this essay we will describe the definitions of learning on computers or devices, so that it serves and other specifications.
DEVELOPING
Learning is something related to the topic that is detailing you, many times we ask ourselves. What is learning?, A general definition related to learning is: as a process that is obtained.
Automatic learning is the branch of artificial intelligence, which allow to develop different techniques that allow computers or devices to learn, computers find algorithms and heuristics to be able to show the data either in a program.
Automatic learning is constituted by an amount of applications, which have search engines, medical diagnoses, for the fraud for the use of credit cards, analysis in stock markets, for DNA sequences, recognition in natural language,Games and Robotics.
MODELS
The models serve to automatically solve a specific task.
Geometric models: are constituted by one, two or multiple dimensions. A limit of the linear decision is constituted by w*x = t, where the w is a vector, x is arbitrary and t is the threshold of the decision.
Probabilistic models are those that determine the distributions of the probabilities of a function with certain values.
Logical models: they are responsible for transforming and expressing each and every of the probabilities into rules through decision trees.
These models are classified in two parts of grouping models and gradient models. Grouping divides the space space into groups. The gradients, as the name indicates, shows a gradient in the difference of each instance.
Types of algorithms
- Supervised learning: this produces a function that makes a connection between the inputs and outputs of the system.
- Non -supervised learning: where the process is done only by the entrances to the system. The system has to recognize each of the patterns to transfer entries.
- SEMI-SUPERVISED LEARNING: It refers to a combination between the supervised and non-supervised algorithm.
- Reinforcement learning: they have the ability to learn by visualizing in all the directions of the world and with a set of information in both directions (from the planet to the machine, and from machine to the planet).
- Multitary learning: In the case of facing an already seen problem they use the processes that you have already learned throughout the period.
CONCLUSION
In my opinion, computers with automatic learning have served as great help for humanity since they have facilitated life doing complex processes without the need for human being to intervene such as: the ability to store information quickly, recognizeEach of us for the process of promoting products for companies and organizations.
There is no doubt that these tools with all the algorithms and models that we have seen with more time will continue to evolve and satisfy customers generating more profits to all companies and organizations.
Learning in the tools have been a fundamental pillar since they have the ability to learn based on experience.
Leave feedback