Man vs. Machine
Game of chess inspired human to invent Artificial Intelligence. The history of chess goes back as long as that of artificial intelligence’s. Centuries before the invention of the computer, the dream of the “Thinking Machine” was linked to the game. It all began with a deception.
The first artificial intelligence (AI) was a lie. 1769 was the year when the world was introduced to a machine that could play chess. The seemingly thinking Android wore Turkish clothes and played chess. It was named “Chess Turk” and it was made by Wolfgang von Kempelen.
The machine consisted of a mechanical man dressed in robes and a turban who sat at a wooden cabinet that was overlaid with a chessboard. The Chess Turk was designed to play chess against any opponent game enough to challenge him. Only the Turk was not an autonomous machine; they had put a man inside of the device!
Georg Gläser and Ernst Strouhal from the University of Applied Arts Vienna wrote a paper “Kemplen’s chess playing pseuodo-automaton and Racknitz’ explanation of its controls 1789”. Explaining Kemplen’s machine, Ernst Strouhal says that if a machine could master the game, it would be a proof that human intelligence can be evenly matched. More than a hundred writings were published following that in which wildest theories were presented as to how the Chess Turk was made. One such reasoning was by magic; the possibilities were countless based on how a small person in the box was able to remotely control and influence the moves via magnetism.
Kempelen always presented the chess Turk only as a magic trick, knowing full well that he would not stand up to scrutiny. After Kempelen’s death the mechanic Johann Nepomuk Mälzel bought the unit and took it on a tour of Europe and United States. Edgar Allen Poe saw the machine and unmasked the falsification in 1836 in a long analytical essay.
When a machine had internalized the principle of playing chess, it had to know all the possible combinations. But the only problem was that the world wasn’t familiar with a thing called the “Shannon Number” then.
To know this, the world had to wait well over one hundred years. In 1950, a mathematician Claude Shannon published his research “Programming a Computer for Playing Chess” in which he estimated the number of all possible chess moves. He based his calculation on a logical approximation that each game has an average of 40 moves and each move a player chooses between 30 possible moves. That makes a total of 10(120) possible games. This number is known as the Shannon Number.
In the following years, the computer strength developed exponentially. 1977 came with the Fidelity Chess Challenger 1, the first commercial chess computer on the market. In the eighties, the devices found their way into many households. Chess programmes for home computers were widespread. In this era, “Battle Chess” was quite popular.
Even in movies, chess was used to prove the superiority of machines. In Stanley Kubrick’s 2001: Odyssey in space, supercomputer HAL 9000 defeated astronaut Frank Poole easily. The Nexus-6 replicant Roy Batty in Blade Runner surpasses its creator Tyrell. On the other hand, Data - the android from Star Trek - loses sometimes a game of 3D chess.
Part of the fiction became reality in the nineties when Deep Blue, a chess-playing computer developed by IBM, defeated Garry Kasparov. Deep Blue won its first game against a world champion on Feb 10, 1996, albeit Kasparov won three and drew two of the following five games, thereby eventually defeating Deep Blue by a score of 4-2.
In recent years, chess has nearly lost its relevance as a testing ground for artificial intelligence. This was assured by the realization that chess is mainly a matter of practice and not necessarily one of intelligence. But this could change now as a programme called “Giraffe” is claimed to have mastered chess in just 72 hours, and more interestingly is game enough to beat anyone who dares challenge it.
Giraffe is designed to play chess in more humanlike ways. Previously, AI programmes used brute-force method, an effective but time-consuming process, which scans millions of possible moves until it finds the best option. Giraffe takes the shorter route. It can identify the moves that are not applicable on the given stage. By eliminating them from its decision-making process, it saves time and energy.
“Unlike most chess engines in existence today, Giraffe derives its playing strength from being able to evaluate tricky positions accurately and understanding complicated positional concepts that are intuitive to humans, but have been elusive to chess engines for a long time,” according to the research paper uploaded to the pre-print site Arxiv.
Despite Giraffe’s success at chess, it was created primarily to showcase the abilities of computer learning and its potential to surpass that of humans’.