Playing chess (pt 1) using human level artificial
intelligence
Note: To make this website free to the public
please click on an ad to support my sponsors or you
can make a tax-deductable donation using Paypal
(click on the donation icon on the left).
Planning programs and
heuristic searches are used to build AI programs to
play chess. IBM's deep blue project is one example of
a planning program that can play chess. Human beings
play chess in a very different manner.
This video depicts a
robot playing chess using human level artificial
intelligence. The video is silent because I wanted
the viewers to see what the robot is thinking when he
is playing chess. This video has two parts (this is
part 1).
The AI chess program
(classified in expert programs) can predict hundreds
of future moves for each chess piece (per second).
The software uses heuristic searches to find the best
future pathway to take to beat your opponent. On the
other hand, a human being can only predict 1 or 2
future moves for one chess piece (per second). That
proves planning programs can outsmart human beings in
the game of chess. This is one of the reasons why
IBM's deep blue project was able to beat the best
chess player in the world.
The problem with
expert programs is that they can only play one type
of game. An AI chess program can only play chess, a
checkers program can only play checkers, a robot car
can only drive a car. The reason human level
artificial intelligence is superior compared to
expert programs is because the robot can do "any"
human task. Not only can this robot play chess, it
can play checkers, monopoly, or any board game. It
can also drive a car, fly a plane, cook in a
restaurant, clean a house, or do any human task.
In the video, the
robot has very little experience in playing chess and
he only has a basic understanding of rules and facts
on the game. He really doesn't use any strategies in
the game. He makes moves based on the current
situation. He does imagine future moves, but it takes
the robot several seconds to imagine one future move
per chess piece and he hand selects which pieces to
predict.
The chess game lasted
for about 10 minutes, but the video is over 20
minutes long and the reason is because I slowed down
the frames. I needed more time to explain to people
how the robot thinks while making decisions in the
chess game. As you can see at the ending, the robot
does indeed beat the opponent, which proves that this
robot has human level artificial intelligence.
The game the robot is
playing is named, chess and is played on the Sega
genesis console.