Playing T2 using Universal 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).
Universal Artificial
Intelligence is one software program that can do any
human task. It can drive a car, fly a plane, play any
videogame, play any sports game, etc. My UAI doesn't
use: machine learning, semantic networks, decision
trees, relational graphs, bayesians network, backward
and forward logic, genetic programming, language
parsers, grammar rules, rule-based systems, common
sense systems, etc, etc. All information is learned
from teachers in school and machine learning is not
used.
This video
demonstrates how a robot soldier thinks while playing
a first-person shooting game. There are no sound in
this video because I wanted the viewers to focus on
the robot's thoughts while playing the game.
What I wanted to show
was that the robot's brain can form a computer
program that does things over and over again. For
example, there are a few rules this robot is actually
following during the game: 1. shoot machines. 2.
reload weapons. 3. shoot machines based on threat
priority. 4. manage multiple tasks. These rules are
followed over and over and over again throughout the
game. Knowledge of shooting games generated these
rules to play the game. Also common sense knowledge
generated these rules to play the game.
This computer program
in the robot's conscious is adaptable as well. The
robot has the ability to change the functions in the
computer program. For example, if a military officer
gave an order to not shoot machines, but instead,
shoot humans, the robot's brain will change the
computer program to include the new rules. Although
this is a hypothetical case, it shows the robot can
change the computer program in its mind, based on the
current environment.
The robot is
constantly making decisions during the game (every 2
seconds). Within those 2 seconds, the robot is
identifying objects, performing logic, analyzing the
situation, and deciding on what actions to take. And
every decision the robot make is based on the current
situation and actions previously made are changed,
deleted, or remain the same. I try to show what the
robot is thinking while making decisions. In the
video, I repeatedly interrupt the gameplay to show
viewers how the robot makes decisions and what
actions the robot will take.
Also, the robot's
conscious creates strategies that it can use to
defeat enemies in the game. The robot's conscious
keeps track of which strategies work and which didn't
work. It's brain will organize and select optimal
strategies to play the game. Prior knowledge or
creative thinking generate these strategies.