Garry Kasparov: Watson Is A Kid’s Toy
By Daniel Long
February 24, 2011 at 12:46 PM
World Chess Champion Garry Kasparov knows a thing or two about defending his pride against game playing computers. He also has a thing or two to say about Watson, that Jeopardy beating robot that nobody likes to invite to parties.
His comments, published in The Atlantic are not surprising and certainly a tad bitter when you consider IBM’s Big Blue humiliated the chess player in a six game defeat in 1997. Basically, his comments go something like this: Watson is a child’s toy, all these computer tournaments should be disregarded and the technologies have little scientific value. No, definitely no sign of bitterness at all.
Kasparov is quick to dismiss the technological achievements of Watson’s game winning powers, comparing its computational brain to that of an 18th century wind-up tool, telling the Atlantic:
“If the result—the chess move, the Jeopardy answer—is all that matters, it’s a success. If how the result is achieved matters more, I’m not so sure. For example, Deep Blue had no real impact on chess or science despite the hype surrounding its sporting achievement in defeating me.If Watson’s skills can be translated into something useful, something groundbreaking, that is the test. If all it can do is beat humans on a game show Watson is just a passing entertainment akin to the wind-up automata of the 18th century.”
Source: http://www.gizmodo.com.au
I have been a huge fan of Kasparov but i have to disagree with his assessment of Watson. Many people do not quite understand how difficult a task it was that Watson accomplished. Understanding the nuances and puns of human language is very difficult for computers to do so it is a tremendous success for the IBM team to program Watson to play and dominate the game. If Watson had won on a simple game show such as Who wants to be a millionaire where simple direct questions are asked and the contestant is given 4 choices and 15-45 seconds to answer then I would not have rated Watson’s performance highly. However on a show like jeopardy where the questions are difficult and contains subtle “human” hints and Watson provides the correct question then I think it was a monumental and historic performance. Especially considering that Watson was competing against the two greatest jeopardy players of all time and he was finding most answers in 3 seconds.
When i first heard about this contest i thought that there was no way a computer can compete at a competent level on a difficult show like jeopardy against ken and brad. However i was greatly impressed and sat with intense amazement at this “game changing” computer task. Even though Watson had an unfair buzzing advantage it does not take away from this amazing computer achievement.
It is very easy for a computer to search through a list of documents and return a list of some that might be useful to answering a question by matching words. The obstacle lies in the computer’s ability to comprehend what is being asked and then parse through various documents and provide the correct answer. Many years from now we will all be looking back at this historic performance as the ground breaking moment that leads to a computer that can understand all the nuances of the human language and can be more useful in our everyday lives. Watson’s grandchildren will be able to fully interact with us and be able to understand exactly what we are asking.
I do agree with Kasparov that deep blue probably did not accomplish much since it used a pure brute force method in defeating Kasparov in the second match. Im sure that many of the top engines today like Rybka and Houdini running on limited computing power would defeat Deep Blue and its army of processing power.
I rate this performance by watson at least 100 times (im not exaggerating) more impressive as the performance by deep blue. I really do hope that Kasparov and the human population fully realize the magnitude of this achievement and its implications for the future of human computer interactions.
“For example, Deep Blue had no real impact on chess or science despite the hype surrounding its sporting achievement in defeating me.If Watson’s skills can be translated into something useful, something groundbreaking, that is the test. If all it can do is beat humans on a game show Watson is just a passing entertainment akin to the wind-up automata of the 18th century.””
Come on Garry, everyone in computer research know of the classified nature of Deep Blue and Watson. Chess has always been used to measure the strength of very special computers.
What were the real world applications?
That is classified information my chess playing comrade.
I can give you a hint.
The world has not been dominated by Russia because of Deep Blue.
It has been suggested that Watson was designed to strategize through collecting satellite data very quickly so that it can quickly evaluate the point of atmospheric reentry, counter solution for tracing missiles back to their origin, and the pending local impact of incoming multiple war heads. Humans cannot operate a missile defense shield effectively, it has to be done by thinking machines. This is just a guess.
Chess is life. Chess is death. Chess is war.
Sore loser?
I’m not sure what Kasparov’s point is. Isn’t he aware that computers have transformed the lives of all people (partly for worse, but generally for much better) over the last 50 years?
I agree with Kasparov’s point, and not the article’s author. What matters about Watson is where it leads. Is this a real breakthrough in programming that can be extended to solve real world problems? I think so, but if not, the Kasparov is right and it’s just a toy.
Why should I listen to the views of Kasparov?
Stef
I’m more with Garry on this one…
Granted it’s a wonderful programming achievement, but will better expert machines make my health care cheaper…make it easier to understand other people …not so far.
Computerization has made one momentous change…everyone’s now on an electronic leash. I’m sure that’s what we wanted.
He needs to clarify what he means by his statement the result isn’t as important as to how the result came about.. does it really matter … its about getting the job done… who cares how the result was obtained.. every computer will use a different means to get the result.. what is important is that it was able to get the result it set out to accomplish..
if we look at the different vchess programs each of these programs uses a different way of obtaining there results.. the progrmaming is sometimes radically different form the next program yet they all play at a very high level .. Alot higher than any human can play..
furthermore isnt chess memory a huge advantage to have when your playing a chess match…having the ability to recall certain situations gives you a huge advantage in a game… Isn’t this what this program is doing.. storing information that its using to recall and win a game.. thats half the battle right there in a chess match..
the top computer programs use huge database tables for the opening and end games so that they can play nearly perfect chess… the only places where they differ is probably in the middle games…he needs to clarify his statements if he wants it to make any sense..
Great commentary wolverine.
Kasparov is full of it. As to chess, surely he would admit that computer programs have had a massive impact on human players’ preparation, game analysis, endgame theory, opening theory, and writing. The implications of Watson are simply breathtaking – and I certainly hope NOT for missile defense. Imagine how useful a computer will be that can take a list of an ill person’s symptoms, ask questions, recommend tests to be performed, and suggest possible diagnoses. It’s literally like something out of Star Trek.
A good post Frederick. You are one of the few persons who seems to truly understand the difficulty of Watson’s feat and its implications and potential on the future of mankind. Theres a great episode of NOVA about the journey and challenges of Watson and the viewer will indeed get a greater understanding and perspective of Watson’s incredible achievement.
To anonymous who asked if Watson will make your health care cheaper, the answer is yes. Watson is currently working in a hospital in new york and im sure that within a few years he will become an invaluable tool for helping doctors diagnose patients. Imagine if theres a Watson in every hospital who can assist with quick diagnosis and recommend tests to run to confirm the diagnosis. Watson will be able to ask the patients relevant questions and understand when they explain their symptoms. Watson of course is not at that level yet but given the extraordinary nature of his recent achievement im very hopeful and excited to see what else he will accomplish.
The beauty of a computer program is that when it it written once it can easily be duplicated and transferred to many machines thus making costs cheaper. The initial versions of Watson would indeed be expensive much like the first home computer which cost more than a house but it will get significantly cheaper with more demand.
Watson’ next task will be learning to master the challenges of helping doctors in hospitals. Watson can also quickly incorporate new studies and treatments in its database since it is almost impossible for doctors to keep up with the high volume of new information especially given their extended hours.
Watson’s functions would not end there but rather his grandchildren would be able to directly interact with us and help us with everything from common everyday activities to financial applications to weather applications and defensive applications to teaching applications to elderly care applications. The capabilities are truly limitless and the future generations will regard Watson as the father of human computer relationships. Watson’s feat will be as revolutionary as the achievements of Sir Issac Newton and Albert Einstein.
Not bad for a child’s play toy
I want be man.
It is about who is funding the development of these computers. Much of IBM’s development budget comes from the US Government. The US Government doesn’t give a damn about health care. It cares about bombs and dollars. These newer computers will help the Government move both bombs and dollars faster.
Russia is smart. Let the Yankee’s develop the computer and send FSB spy women to steal the design.