What is IBM Deep Blue?

What is IBM Deep Blue?

IBM Deep Blue was a chess-playing expert system run on a unique purpose-built IBM supercomputer. It was the first computer to win a game, and the first to win a match, against a reigning world champion under regular time controls. The development of Deep Blue began in 1985 at Carnegie Mellon University under the name ChipTest. It then moved to IBM, where it was first renamed Deep Thought, then again in 1989 to Deep Blue.

Deep Blue was capable of examining about 200 million positions per second, enabling deep search within standard time controls. It first played world champion Garry Kasparov in a six-game match in 1996, where it lost the match by a score of 4-2. After an upgrade, it defeated Kasparov in a six-game rematch in 1997 by winning two games, drawing three, and losing one.

Deep Blue was a combination of special purpose hardware and software with an IBM RS/6000 SP2. Its final configuration used 256 processors working in tandem, with an ability to evaluate about 200 million chess positions per second. The underlying technology of Deep Blue advanced the ability of supercomputers to tackle complex calculations and influenced later work in high performance search and optimization.

Deep Blue's victory marked an inflection point in heralding a future in which supercomputers and artificial intelligence could simulate human thinking. Despite the controversy that followed its victory, the development of Deep Blue was more evolutionary than revolutionary, filled with intellectual challenges and improvements based on experience. Since its seminal victory in 1997, chess-playing computer programs have built upon Deep Blue's developments to become even more advanced.

What was the name of the first chess computer to beat a world champion?

The first chess computer to beat a world champion was Deep Blue. Developed by IBM, Deep Blue made history in 1997 when it defeated the reigning world chess champion, Garry Kasparov, under regular time controls. This marked the first time a computer won a match against a world champion in a formal setting.

Who invented Deep Blue?

IBM's Deep Blue, the first computer to win a chess match against a reigning world champion, was developed by a team led by Feng-hsiung Hsu and Murray Campbell. The development of Deep Blue began in 1985 at Carnegie Mellon University under the name ChipTest, with Hsu as a graduate student. Hsu, along with Campbell and other colleagues, later joined IBM to fully develop Deep Blue. Other key contributors to the project included Thomas Anantharaman, Jerry Brody, and C. J. Tan. Grandmaster Joel Benjamin also played a significant role in the development of Deep Blue, assisting the team with preparations for the matches against Garry Kasparov.

What was the architecture of Deep Blue?

IBM's Deep Blue was a chess-playing computer system that combined both hardware and software components to achieve its capabilities.

Hardware Architecture

Deep Blue's hardware was based on a unique purpose-built IBM RS6000/SP supercomputer. In its final configuration, it used 256 processors working in tandem, with an ability to evaluate 200 million chess positions per second. The system was a hybrid, combining general-purpose supercomputer processors with chess accelerator chips.

The single chip design developed by Feng-hsiung Hsu formed the backbone of Deep Blue's hardware. This chip was designed as a 32-bit device with a 17-bit address space. One of the possible addresses, when written, initiated a search from the current position on the chip, usually for 4 or 5 plies beyond the software search depth.

Software Architecture

Deep Blue's software was designed to carry out part of a chess computation and then hand off the rest to the chess accelerator chips. The software was based on an alpha-beta search algorithm, a type of search still used by many chess engines today.

The software architecture of Deep Blue included several key components:

1. Move Generation — This function was responsible for generating all possible moves for a given chess position. The move generator was an 8 X 8 array of combinatorial logic acting as a silicon chessboard.

2. Evaluation Function — This function was used to evaluate the desirability of a given chess position. It was composed of fast evaluation and slow evaluation components.

3. Search Control — This function controlled the depth and breadth of the search through the game tree.

4. Extendability — This allowed for the addition of other functions like generation of checking, check evasion moves, and developing certain kinds of attacking moves.

Deep Blue's software also made use of heuristics and optimizations specific to the structure of chess. For example, if a move led to the player's king being in checkmate, the algorithm would not look any further down that path of the game tree.

The development of Deep Blue's architecture was a significant milestone in the field of artificial intelligence and computer science, demonstrating the potential of AI in complex problem-solving tasks.

What was the approximate cost of Deep Blue?

The exact cost of developing IBM's Deep Blue is not publicly disclosed. However, considering the complexity of the project and the resources involved, it was a significant investment. Comparisons to later projects are difficult because scopes and budgets varied widely across programs.

How did deep blue's victory over garry kasparov impact the development of ai?

Deep Blue's architecture, widely applied in industries like financial modeling, has provided valuable insights into designing computers for large-scale data analysis. While its AI approach, based on brute force computing power and hand-coded rules for chess, differs from today's dominant machine learning techniques, Deep Blue's victory remains a landmark moment. It sparked debates about the future of human-machine relationships and showcased the potential of AI in diverse applications, including medical diagnosis and financial analysis.