Reinforcement learning, whereby an agent tries to maximize the reward in a “complex, uncertain environment”, is just a computational approach to interactive learning. Such interaction with the environment is how people learn. Instead, using reinforcement learning, humans’ learning process was mimicked by studying an impressive number of chess games. The AlphaZero algorithm did not try to use the brute force of computing power to identify as many moves as possible on the chessboard. Significant progress in artificial intelligence–related fields was made only 20 years later when AlphaZero won a chess match in 2017 against Stockfish, one of the most powerful chess engines ever created, after a self-learning process of only 4 hours. In the following years, computing power advanced to the point where even the best chess players had no chance of defeating a modern chess engine, as has been previously stated. However, this victory was mainly achieved by brute force, since the Deep Blue program searched millions of positions per second to play chess at a slightly higher strength (the program won the 6-game match with Garry Kasparov with a difference of only 1 point), so it was probably more impressive that Kasparov, using human intuition, could still be almost as strong as a computer searching millions of positions per second.
One of the most famous human versus machine events was the 1997 victory of Deep Blue, an IBM chess software, against the famous chess champion Garry Kasparov. The progress of artificial intelligence, highlighted by strategic games, has affected many other areas of interest, as has already been seen in recent years. The developments in artificial intelligence for chess have advanced beyond gaming, changing the way machines and humans coexist. Chess has inspired artificial intelligence progress for decades.
Context: Artificial Intelligence Progress Displayed in ChessĪrtificial intelligence is undergoing a revolution, but at the same time, it has caused revolutionary changes in the world.
0 kommentar(er)
