Revenir au site

A brief history of algorithms since the 9th century

· Part I

Anna SKVARCHYNSKA

It all started in the 9th century. The word “algorithm” comes from the name of the great Central Asian scientist Abu Abdullah Muhammad ibn Moussa al-Khwarizmi. Of many mathematical works of Al-Khwarizmi, only two have reached us - algebraic and arithmetic. The second book was long considered lost, but in 1857 in the library of Cambridge University its Latin translation was found. It describes four rules of arithmetic operations, practically the same ones that are used now. The first lines of this book were translated like this: "I said Algorithms. We will praise God, our leader and defender. " So the name Al-Khorezmi passed into Algorithms, from which the word algorithm appeared. The term algorithm was used to denote four arithmetic operations; and in this sense it appeared in European languages.

broken image

Then around 1360, Scientists began to apply it not only to purely computational, but also to other mathematical procedures. French philosopher Nicole Oresme wrote a mathematical treatise "Algorismus proportionum", in which he first used degrees with fractional exponents and actually came close to the idea of ​​logarithms. When the abacus was replaced by the so-called account on the lines, numerous manuals on it began to be called "Algorithmus linealis" - the rules of the account on the lines.

In 1684, Gottfried Leibniz, in the “Nova Methodvs pro maximis et minimis, itemque tangentibus ...,” first used the word "Algorithmo" in an even broader sense: as a systematic way to solve the problems of differential calculus.

Then in 1759, Leonard Euler used the word “algorithms in a sense very close to modern in one of the works - "Using a new algorithm to solve the problem of Pell" ("De usu novi algorithmi in problemate Pelliano solvendo") –algorithm as problem solving method.

The word "algorithm" can be found in the popular pre-revolutionary Encyclopedic Dictionary of the Granat brothers, and in the first edition of the Great Soviet Encyclopedia, published in 1926. The meaning stays the same: as a rule, according to which this or that is fulfilled from four arithmetic operations in the decimal number system. However, by the beginning of the 20th century for mathematicians the word "algorithm" already meant any arithmetic or algebraic process performed according to strictly defined rules.

During the year 1936, The precise definition of the concept of the algorithm made it possible to prove the algorithmic insolvability of many mathematical problems. Alan Turing described a theoretical machine that could solve problems by following coded algorithms.

In 1950, his next paper was called “Computing machinery and intelligence”. Since emergence of computing machines – the idea of algorithms and smart machines were often mentioned together, and although some visionaries of Turing on AI have not been implemented yet, more and more recent achievements prove that soon it will be the new reality.

Few years later, in 1956, two significant events took place: “Logical Theorist” and the Dartmouth AI Conference. “Logical Theorist” was a working Ai machine, created by AL Newell and Herbert Simon. It could deduce geometric proofs, with a programming language and direct put in information.

Dartmouth Conference was the largest gathering of that time scientists on the new disciplinary. Several topics were discussed, such as automatic computers, how can a computer be programmed to use a language, neuron nets, the theory of the size of a calculation, self-improvement, abstractions, randomness and creativity.

In 1958, McCarthy has introduced LISP –innovative language for AI programming.

A year later, Minsky explores intelligent machines theories in “Computation: Finite and Infinite Machines”.

In 1963, Thomas Evans presents “ANALOGY” – a program, capable to SAT-level tasks by using semantic rules for information interpretation.

In 1961, a first industrial robot working for GM introduced, UNIMATE, and a robotic hand MH1 project.

broken image

In 1966, Vision Project was an attempt to teach AI transform visual data into information. This initiative has appeared ahead of its time and thus have not succeeded.

first conversational machine, ELIZA. It simulated a psychologist by following behavioral patterns.

A year later, LOGO is a new programming language, used for primitive robot programming.

In 1979, Mycin, the infectious blood diseases diagnosing & recommending system has outperformed members of Stanford medical school. However, it has never been used in practice due to many ethical questions of responsibility.

In 1982, HEARSAY is a speech understanding machine, able to transform input sound signals into information, process and decode.

In 1997, Deep Blue, IMB-built machine, has defeated Garry Kasparov, world chess champion in “Fight of 20th century”- symbolical match between human and artificial intelligence.

broken image

Recently, in 2005, US military invests in war robots, producing BigDog- the first of the serie of robotic animal. Later, US has deployed over 2000 bomb disposal robots in the fields of Iraq and Afghanistan.

Check out this video on BigDog evolution by Boston Dynamics:

In 2008, Google introducted of voice recognition software on the users Iphones. Since then, Google has introduced several improved and revolutionary version of Go, programming language, one of the most commonly used nowadays.