Playing tetris 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).
IBM's deep blue project
can only play chess, an autonomous car can only drive
a car, an AI checkers program can only play checkers.
A universal artificial intelligence is one software
program that can do any human task. It can drive a
car, play chess, play any videogame, cook in a
restaurant, clean a house, fly a plane, write
software programs, etc.
So far I have made
videos on playing chess, playing videogames, driving
a car, and doing math equations. I stated in my
books/patent apps/website, that my robot can do any
human task. This video shows the robot playing tetris.
In the video I show how the robot thinks while he's
playing the game.
Tetris is a puzzle
game that require the player to stack up blocks to
form stacked rows. Humans learn how to stack up
blocks and solve puzzles starting from a very young
age. Stacking blocks so the pieces fit are lessons
learned in Kindergarten and in grade school. When the
robot is playing tetris, he has to tap into knowledge
about block stacking or puzzle stacking. He has to
know that this shape fits into that shape.
In addition to
knowledge about piecing together puzzles, the robot
has to have strategies to play the game. He has to
have objectives, and recursive objectives and so
forth. These objectives and strategies are discovered
through either trial and error or from instruction
manuals. The robot can read strategies from a
videogame magazine or he can discover strategies
based on trial and error.
Also, making
decisions about where should blocks be stacked and
conflicts of interests are learned from teachers in
school. By the time the robot is at age 10, his brain
contains self-learning pathways or self-adaptive
pathways. He has the knowledge to teach himself the
best strategies to play tetris. For example, no
teacher has taught the robot how to play tetris. He
used his own intelligence to find out what the game
is about, how to play the game, what are the best
strategies in the game, what decisions to make, and
so forth.
The truth is that
playing tetris is a very complicated thing to do. The
video only shows a general way the robot thinks as it
is making decisions in the game. The internal
instructions are much more complex.
This robot doesn't
use planning programs or heuristic searches. This is
not an expert program that was designed specifically
to play tetris. This robot can play chess, play
checkers, play monopoly (or any board game), drive a
car, fly a plane, play any videogame, or do any human
task. Programmers don't have to change the robot's
brain for each human task.