Have robots overtaken us? “Artificial intelligence can be far more intelligent than man”

Robots were designed to prevent humans from performing dangerous tasks and routine tasks. From this Czech word meaning ‘forced labour’, it was first used over 100 years ago in Karel Capek’s ‘RUR (Rossum’s Universal Robots)’, and has since entered our daily lives.

Recently, a group of scientists from Kyoto University created an artificial intelligence system to put in a robot called Erica. it’s android able to laugh at the right time Furthermore, from a conversation, the algorithm learned the basic characteristics of social and joyful laughter aimed at reflecting it in appropriate situations.

To carry out this process, the researchers collected training data from more than 80 conversations between university students, so the information was used to develop a machine learning system Able to decide whether a robot can laugh,

Aseco Spain’s data scientist Carlos Alberto García explains for 20 bits that Robots identify the right time to laugh due to trial and errorSaying that the device “obviously doesn’t make sense, but will know it’s time to laugh about the programmed algorithms and its machine learning capabilities.”

This innovation may be a bit vague among users, but what exactly is the purpose of Erica? Changing people, showing empathy, or being a mentor?

Santiago Ferrer Jover, head of industry at ASECO Spain, explains that “the intention is not to make the robot understand jokes, but show sympathy, This is very important so that humans feel less apprehensive when interacting with robots. If the robot is friendly, people may feel more relaxed and open to interact. you have to try make them friendly Possible”.

It is in the shape of a cart.

What data do robots store?

loss training information They are necessary to teach androids to perform certain actions, so what kind of information is collected to make them similar to people?

García responds that “basically ConversationsEither in audio or text. It is this type of information that provides the most data about humans and that is easy to assemble and analyze, ‘Natural Language Processing’ is the area of ​​study of artificial intelligence that has developed the most in recent years, and which is constantly giving rise to new advances”.

“The study of human speech is extremely promising for many interaction systems such as virtual assistants, information and guidance points, leisure and entertainment, therapeutic implications for chronic pathology in hospitals and residences. Robots are here to serve us and they are becoming more effective“, says the data scientist.

It usually takes several weeks or months to process this type of information.However, there are transformers that speed up the whole process (it can still take several hours).

The Gogo co-founder says his robots won't replace riders.

they are able to manage

NetDragon Websoft (Chinese mobile game company) appointed Android CEO Working With Artificial Intelligence, Tang Yu, as he is called, held the position of general manager at the brand’s main subsidiary (Fujian NetDragon), making decisions during daily operations and ensuring a proper work environment for employees.

Ferrer explains that a robot can make decisions when sufficient information is provided, Similarly, he points out that “they analyze and consider a large number of variables at the same time, such that they are not carried away by their intuition or their feelings because they believe only in the data”.

Czech journalist and philosopher Karel Capek coined the term 'robot' in his Rossum Universal Robots (1920).  In it, a company used advanced biotechnology to mass-produce workers without spirit or emotion so that they could be over-exploited.

If they make decisions, can they make people smarter?

“If recent projects like AlphaGo have taught us anything, it’s Artificial intelligence may be smarter than humansin very specific tasks,” says Garcia.

“Alphago was able to defeat the champion of the sport and make new plays that had not been taught before, and even no one knew, however, if we asked him to tie his shoes, He will fail miserably. The problem is not to outperform a human at a task, but to generalizeAnd I think we’re decades away from that.”

Advances in neural networks and quantum computing aimed at computing capability would allow extraordinary advances to offer Process millions of data in seconds,

Sign up for our newsletter and receive the latest tech news in your email.

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button