contador javascript Saltar al contenido
Contact : alejandrasalcedo0288@gmail.com

Artificial intelligence: from Kasparov to Sedol, two ways to face defeat at the machines


Garry Kasparov faces Deep Blue in the legendary chess games Source: Reuters

A frightening milestone in the history of mankind. "As value Kasparov su

defeat against Deep Blue

. And that had not yet happened. He said this before in a meeting with several journalists who attended EL PAS. He also stated that he was convinced that this dreadful victory will occur at some point. "What I try is to postpone it as much as possible." A few days later, on May 11, 1997, I lost. And I adjusted it regularly.

First accuse
        
        
IBM

        

 , responsible for the intelligence of the rival machine, of cheating:

"Deep Blue showed signs of intelligence

and made decisions that human beings made by intuition. "Then he refused to admit failure." Let no one interpret this as a definitive defeat of man before the machine. The competition has only just begun. "In fact, Leontxo Garca, who was there covering the duel for EL PAS, attributes what happened in the end to the" nervous system "of the genius." The mistake I made was not typical of a teacher of chess, "recalls the expert, who also rules out that IBM cheated.

However, defeat ended up penetrating the collective imaginary as the beginning of the end. Since then, defeat by defeat, we have been adjusting our expectations about the games that artificial intelligence can win and those that are not within reach. Kasparov has already assumed the part that touches him of the "awful milestone". More than twenty years later,

in an interview with Wired,

 He recalled that day in these terms: "I have made peace with the subject. In the end, the game was not a curse but a blessing because I was part of something very important. 22 years ago I would have thought something different, but things happen. We all make mistakes. We lose. And the important thing is how we deal with our failures, with negative experiences. " Now Kasparov looks at inhuman intelligence with other eyes. "1997 was an unpleasant experience, but it helped me understand the

future of human-machine collaboration

".

The other face


AlphaGo defeated Lee Sedol three times, but the specialty champion managed to beat the Google system in the fourth of the five agreed matches

AlphaGo defeats Lee Sedol three times, but the champion of the specialty managed to beat the Google system in the fourth of the five agreed matches

This change in perspective has taken more than 20 years to arrive but contrasts with the position of

Lee Sedol

, whoever it was

Go world champion

. In 2016, the Korean lost against artificial intelligence developed by

Deepmind

(Google) to master this strategy game,

Alphago

. And three years later he got off the professional Go lite, unable to overcome his failure at the algorithm. "With the debut of artificial intelligence in Go games I have realized that I am not at the top, although I try to be number one through frantic efforts," he conceded to the Yonhap news agency.

Kasparov described himself at the end of last year as "the first intellectual worker painfully defeated by a machine". Lee Sedol is one more in that wake to which two champions of the strategy videogame also joined in 2019

Starcraft

. "Even if I become number one, there is an entity that cannot be defeated," said the newly retired Go player.

Sedol only won one of the five games in which he faced the algorithm on March 13, 2016. And yet, his only victory can be considered a success: no other human Go player has managed to win against AlphaGo. "I rarely read comments about me on the news on the Internet. But I was curious how bad people will be talking about me after my defeats. Unexpectedly, few people criticized me."

Hand in hand with the machine

"We thought we were invincible in chess, Go, Shogi. All these games have gradually been left apart. But this does not mean that everything is lost. We have to find a way to play in our favor," Kasparov prescribes in Wired. The teacher insists that we have to let the machines do what they are best at, not in vain "technology is the main reason why many of us are alive to complain about technology."

The future as he goes through the "combination" of skills, but also demands humility. Medicine is an example: "Someone with decent knowledge understand that you must add little. But a great medicine star try to challenge the machines. That destroys communication."

Magnus Carlsen

, modern chess genius, is already following this path. Instead of antagonizing the algorithm and despair of defeat, he has focused on learning from his playing style. The 29-year-old Norwegian describes

AlphaZero

(an heir system of AlphaGo), like "your hero".

We are all Kasparov

Anyway, the clashes with the intelligence of the machines are no longer a genius thing. And the reflections Gari Kasparov has made throughout the decades are not so far from what we do today.

  • When Google recommends you the best dishes of the restaurant where you just sat:

    "I face an unknown alien who has all the possible information about me."
  • When nobody explains how the algorithm decides:

    "I propose to IBM to print and publish the entire Deep Blue reflection process during each of the six games of this duel. Then maybe we can understand why my opponent has made incredible plays on a computer."
  • When Alexa does not understand the title of the English song you are asking:

    "I have made a typical human error of imprecision. In addition, humans always want more. But I think I must be satisfied."
  • When you leave home with your smile on:

    "Humans know that there are many animals and machines that run more than us and are stronger and more agile. But none of them is smarter and smarter than us."

ADEMS

. (tagsToTranslate) Artificial intelligence: from Kasparov to Sedol (t) two ways to face defeat against machines – LA NACION