The quest for Artificial intelligence - Part 1 - The Deep Blue story

Jottings - Slice of Life - 316 ( The quest for Artificial intelligence - Part 1 - The Deep Blue story )

(Note to my readers: This piece is the first of my intended three-part essay on Artificial Intelligence, and some of its glamorous role models in the last three decades. The essay is nearly 2500 words long and should take about 15 to 20 minutes of reading time. As always, I attempt to write for the general reader, who has the time and curiosity to learn more. I don't expect any prerequisite knowledge in computers to follow this essay, and at the same time, I haven’t diluted the technicality of the language to make it mean something less technical. Popular writing mustn’t presume its readers to be ignorant or lazy. If there are terms that are new, I request the reader to pause, look up the term and proceed.)

In the early months of 1997, within the majestic Equitable center building in midtown Manhattan, occupying the 32nd floor in its entirety, sat one of the most powerful computing machines ever built by human hands. It sat there quietly (unconscious, of course) whirring in its mechanical glory, cooled by a dozen fans, powered by its terrorizing computational efficiency, weighing thirty-two tons in all, containing 32 nodes ( a node is a machine) with each of them equipped with an integrated chip — the heart of a computer — capable of performing more than hundred million computations per second. Together, all the chips working in parallel could churn out results at exponential speeds and efficiency. In building those chips lay the work of outstanding mechanical engineers, informational scientists, software developers, student researchers, and corporate support. The circuitry of each chip was algorithmically built and refined over twenty years of pioneering academic work, all for just one single purpose — to play chess, as well as Humans, do, if not better. This behemoth of a machine was christened “Deep Blue” by IBM ( who sponsored this effort). The prefix “Deep” is derived from the name of the fictional supercomputer that featured in Douglas Adam’s popular book “ The Hitchhikers Guide to the Galaxy”.

For long, the game of chess has been considered the archetypal symbol for Human intelligence. The spectacle of the chessboard with its sixty four squares in black and white; the coins on the board metamorphosing into rival armies as kings, queens, bishops, horses, pawns and rooks; the curious rules,, strategic openings and the rigid movement of the coins across the board, the unbelievable number of documented attacks and defenses; the subtlety of the psychological warfare between players, and above all, the capacity of the human brain to think ahead and visualize the possible state of the coins on the board, based on the current move - makes chess a highly cerebral game which need an extraordinarily nimble brain to store, assess, refactor and compute over long stretches of time. To a large extent, it is muscle memory. In general, Chess is played and relished only by people with an aptitude to retain information coupled with exceptional computational skills. The picture of the man playing chess with his heads bowed down, elbows resting on the table with chins held within the palms, eyes deeply focussed, concentrated, oblivious of the surroundings and completely absorbed in the intellectual battle at hand - has provided irresistible imagery of chess epitomizing supreme intelligence suffused with artful maneuvers. There was something about chess that went beyond the mathematics of it, and culturally, winning a game of chess became a measure of intelligence. Even Claude Shannon, the founding father and arguably the man behind the evolution of information science in the late 1950s worked out a strategy for computers to play chess. Shannon insightfully perceived the deep connection between representing information as binary rules, and the ability to apply them to a game such as chess. But to him, and many others after Shannon, such efforts were merely intellectual experiments without any conviction of immediate consummation. The sheer complexity of predicting the next move, or series of moves after that, involved an algorithmic complexity that humans seem to effortlessly possess, and so difficult for a machine to emulate. Great players “seem to know” what their opponents are thinking, and are able to devise a strategy based on the context. With computers, there is no context, every move has to be tested and evaluated from scratch with no sense of Knowing. But the disadvantage of context is often offset by the relentless computational power and the absence of any conscious attachment to the task at hand. In his wonderful book “ Godel, Escher, and Bach”, the brilliant cognitive scientist Douglas Hofstadter hinted at the possibility that someday machines may be computationally fast and powerful enough to play and possibly win at chess, but he wasn’t in favor of such success. In a poignantly prescient statement professor, Douglas wrote in his book: “What bothers me is the degree to which something incredibly simpler than our brain is starting to be able to do things that we do in surprisingly strong ways. It's taking away from the complexity of what we really are.” Chess represented a unique capacity of the human brain. And to program machines to take over that capacity seemed dehumanizing and demeaning too, in some respects.

It is fascinating that the real race between Man and Machine in chess began as a student research project in the early eighties. When Feng-Hsiung Hsu arrived in the USA from Taiwan in 1982 with a Bachelors degree in Electrical engineering, the only thing he was certain of was his conviction that a machine that could compute faster would be a great advantage in playing chess. His Masters in Carnegie Mellon focused on integrated circuits - those mysterious connections that pulse electric signals through the labyrinthine circuitry of the chip ( getting smaller each day) to perform hundreds and millions of computations. It is the design of the chip and its intricate circuits that power a computer. Moore’s law - the famous prediction that computing power will increase every two years - is all about how many transistors ( channels of communication and computation) can be packed in a tiny microchip. There is a physical limit, but in the eighties, the field was open, and Moore’s law was still active. Hsiu worked with three different teams and projects before IBM adopted him in the nineties to work on Deep Blue. In 1988, a few years before Hsiu joined IBM, “Deep thought”, the machine that Hsiu had helped build earlier had lost to Kasparov, the reigning world champion. Hsiu learned a number of things from that loss. The foremost being that the number of strategic moves a computer had to evaluate was still not equal to the task. Human’s instinctively knew what to play, and what to avoid, but a machine has no such self-conscious knowledge. The specific algorithm designed for chess was called the “Evaluation function”, and its job was to apply brute force to each move. What it means is taking each move as a discreet unit and evaluating all possible options from that point onwards. The machine cannot reject any move based on “experience”. Its incredible efficiency lay in slogging through the million sets of evaluations needed to assess the impact of a single move on the chessboard. Its strength was its relentlessness and tirelessness. When IBM announced the Deep blue challenge, Hsiu knew what needed to be done. He must find a way to pack more computations on a single chip, and the software ( evaluation function) must be fed with sufficient training data. Both of which needed money, effort, and coordination, and IBM was willing to put up the stakes necessary.

In the annals of Chess history, Kasparov is considered the greatest chess player. For twenty years, from 1985 when he started playing the game professionally to 2005 - when he retired from the arena, Kasparov was virtually undefeated. At the age of twenty-two, he beat Karpov ( another Russian) to become the youngest world champion, a position he held for two decades. He was not merely the greatest strategist the game has ever known, but Kasparov’s ability to outthink the opponent by several moves, sudden bursts of unconventional play, and a calculated arrogance that unnerved his opponents. He was a recluse, self-centered, introverted and almost disdainful of everyone else. Like Mohammad Ali in Boxing, Kasparov believed that there was no other in the world who can be his equal. If winning a game of chess is the true measure of intelligence in a machine, then beating Kasparov will be the crown jewel of any such effort. IBM was willing to invest in that challenge. It was a calculated move by IBM. If it paid off, then, once again, IBM would be applauded for ushering in the new wave of Information technology. If it fails, there was nothing to lose. After all, Kasparov was expected to win. In the May 1997 edition of Newsweek, Steven Levy, the renowned journalist, and chronicler of technology persuaded the magazine’s editors to run a cover page featuring Kasparov vs Machine match. Reluctantly, the editors agreed to publish the legendary cover line “The Brain’s last stand” with a picture of a serious-looking Kasparov peering at the reader. Fortunately, there were no celebrity deaths, or political upheavals that week. Therefore the May edition was well received by the reading public, and an active appetite generated for the tournament to follow.

In 1996, when Kasparov played the first version of Deep blue in Philadelphia, he beat the machine easily with a final score of 4-2. The IBM team huddled for the next thirteen months fixing the algorithm, feeding the machine with more refined datasets, and bolstering the computational powered. By May of 1997, Deep blue was a different beast altogether, and a hundred times more powerful and tuned to play chess. In the rematch of 1997, Kasparov would play six games, with all the rules of the game in place, including the time controls. Based on the machine’s advice, a computer scientist from IBM, Stuart Campbell, would move the piece on the chessboard. Kasparov won the first game quite easily in forty-five moves using the Kings Indian attack - a classic chess opening normally employed by champions. Going into game two, Kasparov was on a roll and confident. On the 36th move, Kasparov left his queen exposed for his opponent to take. It was part of the Smyslov variation strategy, and most other players would have taken the queen and played to the script. But Deep blue came up with a very unusual and subtle move that caught Kasparov off guard. It left the queen alone and moved the pawn. In that single move, IBM’s Deep Blue had demonstrated an ability outthink and plan beyond the most accomplished strategist of the game. In an interview after the game( and years later too), Kasparov would isolate that 36th move as his moment of realization and awakening. After that move, Kasparov began to become conscious of his game plan, and little agitated too. Within fifteen more moves, Kasparov conceded the game and give Deep Blue its first victory in the tournament. Kasparov later cried foul and demanded that IBM show him detailed logs of the moves. IBM produced snippets of the logs, but never showed the world the whole piece. Many years after the fire of the controversy cooled down, IBM revealed that in the second game Deep Blue’s 36th move of the game was actually algorithmically flawed, totally unintended, and never part of the strategy. An inadvertent computer error cost Kasparov his cool, and the game itself.

The next three games were drawn, and the outcome of the last game - the sixth would be the decider. Kasparov opened the game with the Caro-Kann defense which would require a sacrifice of the knights by an opponent without any immediate gains. Kasparov reasoned that a machine wouldn’t play that move because it did not lead to any strategic advantages; but to his surprise, Deep blue did make the Knight’s sacrifice as required and willing to wait for a subsequent opening. What Kasparov didn’t know at the time he played was that very morning of the last game IBM programmers had fed the Caro-Kann variations into the deep blue. The machine was prepared for the bait. Kasparov resigned in more twenty moves, and Deep Blue had scored an improbable victory over the champion, and the human mind, so to speak.

In writing this article, I have condensed large tracts of material to make it more accessible to a general reader. The triumph of deep blue in 1997 ushered in a new age of confidence in the world of Information technology. Within a few years, Google would be born, and along with Apple and Microsoft, would transform the social landscape of how data can be processed and used. Kasparov himself practiced and refined his skills against computer-simulated games, but in the end, the very machine that supported his practice humbled his pride. Of course, there was a huge human team behind Deep blue’s success, that trained the machine not merely to be superiorly efficient in computation but a little cunning and deceptive too. For instance, they programmed Deep Blue to slow down on crucial moves, and prolong the act of making the move, to induce a sense of indecision and uncertainty. And on other times, when the opponent makes a crucial move, the machine would react spontaneously. As a machine, a human would expect consistent responses from it, but by deliberately varying the response times, the designers were attempting to wage psychological warfare with the opponent. And that’s the scary part of AI. Somewhere along the line, we painfully realize that machines don’t possess any intelligence or knowledge. It is we who give it direction and “intention”. The other factor that must be borne in mind is that Chess is all about comparing the value of a move in relation to the coins on the board. The rules of the game can be symbolized in mathematical terms and assigning appropriate weights to each move. A robust algorithm can be built that can minutely compute, factor and compares results and patterns from its repository to produce the next move. If a given move builds confidence in the algorithm and increases the chances of a win, it is deemed a numerical victory, otherwise not. Highly trained Human brains - such as Kasparov or Anand’s - are capable of holding and visualizing the outcomes of their moves at least five to ten steps ahead. How the brain does it is still a mystery, and computers are nowhere close to emulating the nimbleness and subtlety of the brain in arriving at such a visualization. But computers make up in computational speed and tireless efficiency what it lacks in the human ability to visualize. Deep Blue is the perfect manifestation of such a machine.

After Deep Blue’s stunning victory, when the euphoria settled down, the important question arose on what really was the achievement here. Did we concede that Machines can be intelligent in a way Humans are, not just in playing chess, but in all other aspects as well? Or was Deep Blue’s victory merely an exaggerated attempt at mastering a specified skill which we seem to effortlessly exhibit with some practice? Is artificial intelligence the right phrase at all to describe what Deep blue had achieved? Or were we simply carried away by what we got the machine to achieve. Broader questions on what it means to be intelligent, or understanding the language in all its nuances, or create intellectual abstractions, or create art - still seemed beyond the ken of AI. The Deep Blue story belongs to the last decade of the twentieth century. The twenty-first century has radically improved upon the capabilities of Deep Blue, and with each passing day, we seem to take over more and more of specialized computational tasks from Human hands. However, we are still very far away from creating a mechanical prototype equivalent to the complexity, intricacy, and elasticity of the human brain. In 2011, IBM unleashed Watson, another significant advancement over Deep Blue, This time around, the machine was equipped to understand human language and respond to clues.

I will write about the Watson story in an upcoming installment.

God bless…

yours in mortality,

Bala