E-Business

a. Evaluate the categories of artificial intelligence and how its implementation may bring betterment to the people.

Artificial Intelligence (AI) is the area of computer science focusing on creating machines that can engage on behaviors that humans consider intelligent. The ability to create intelligent machines has intrigued humans since ancient times and today with the advent of the computer and 50 years of research into AI programming techniques, the dream of smart machines is becoming a reality. Researchers are creating systems which can mimic human thought, understand speech, beat the best human chess player, and countless other feats never before possible. Find out how the military is applying AI logic to its hi-tech systems, and how in the near future Artificial Intelligence may impact our lives. (ThinkQuest, 1997).

There are several topics of AI that be discussed in detail. The topics are:

1. Games playing

In a number of games, computers have enjoyed success that puts them on par or better with the best humans in the world. In some sense, these games are now the past, in that active research to develop high-performance programs for them is on the wane (or is now

nonexistent). These include games where computers are better than all humans (checkers,

Othello, Scrabble) and those where computers are competitive with the human world champion (backgammon, chess).

1.1 Checkers

Interest in checkers was rekindled in 1989 with the advent of strong commercial programs and a research effort at the University of Alberta--CHINOOK. CHINOOK was authored principally by Jonathan Schaeffer, Norman Treloar, Robert Lake, Paul Lu, and Martin Bryant. The structure of CHINOOK is similar to that of a typical chess program: search, knowledge, database of opening moves, and endgame databases (Schaeffer, 1997; Schaeffer et al., 1992). CHINOOK uses alpha-beta search with a myriad of enhancements, including iterative deepening, transposition table, move ordering, search extensions, and search reductions. CHINOOK was able to average a minimum of 19-ply searches against Tinsley (world champion player) using 1994 hardware, with search extensions occasionally reaching 45 ply into the tree. The median position evaluated was typically 25-ply deep into the search.

A notable feature in CHINOOK is its use of endgame databases. The databases contain all checkers positions with 8 or fewer pieces, 444 billion (4 _ 1011) positions compressed into 6 gigabytes for real-time decompression. Unlike chess programs, which are compute bound, CHINOOK becomes input-output bound after a few moves in a game. The deep searches mean that the database is occasionally being hit on the first move of a game. The databases introduce accurate values (win/loss/draw) into the search (no error), reducing the program's dependency on its heuristic evaluation function (small error). In many games, the program is able to back up a draw score to the root of a search within 10 moves by each side from the start of a game, suggesting that it might be possible

to determine the game-theoretic value of the starting position of the game (one definition

of "solving" the game).

CHINOOK is the first program to win a human world championship for any game. At the time of CHINOOK's retirement, the gap between the program and the highest-rated human was 200 rating points (using the chess rating scale). A gap this large means that the program would score 75 percent of the possible points in a match against the human world champion. Since then, faster processor speeds mean that CHINOOK has become stronger, further widening the gap between man and machine.

1.2 Chess

The progress of computer chess was strongly influenced by an article by Ken Thompson that equated search depth with chess-program performance (Thompson, 1982). Basically, the paper presented a formula for success: Build faster chess search engines. In 1996, the chess machine played a six game exhibition match against Kasparov (former world champion). The world champion was stunned by a defeat in the first game, but he recovered to win the match, scoring three wins and two draws to offset the single loss. The following year, another exhibition match was played. DEEP BLUE scored a brilliant

win in game two, handing Kasparov a psychological blow from which he never recovered.

In the final, decisive game of the match, Kasparov fell into a trap, and the game ended

quickly, giving DEEP BLUE an unexpected match victory, scoring two wins, three draws, and a loss. It is important to keep this result in perspective. First, it was an exhibition match; DEEP BLUE did not earn the right to play Kasparov .5 second, the match was too short to accurately determine the better player; world-championship matches have varied from 16 to 48 games in length. Although it is not clear just how good DEEP BLUE is, there is no doubt that the program is a strong grand master.

The notable technological feature of DEEP BLUE is its amazing speed, the result of building special-purpose chess chips. The chip includes a search engine, a move generator, and an evaluation function (Campbell, Hoane, and Hsu, 2001; Hsu, 1999). The chip's search algorithm is based on alpha-beta. The evaluation function is implemented as small tables on the chip; the values for these tables can be downloaded to the chip before the search begins. These tables are indexed by board features and the results

summed in parallel to provide the positional score.

A single chip is capable of analyzing over two million chess positions a second (using

1997 technology). It is important to note that this speed understates the chip's capabilities.

Some operations those are too expensive to implement in software can be done with little

or no cost in hardware. For example, one capability of the chip is to selectively generate subsets of legal moves, such as all moves that can put the opponent in check. These increased capabilities give rise to new opportunities for the search algorithm and the evaluation function. Hsu (1999) estimates that each chess chip position evaluation roughly equates to 40,000 instructions on a general-purpose computer. If so, then each chip translates to a 100 billion instruction a second chess supercomputer. Access to the chip is controlled

...