Generating common sense knowledge (pt
6) 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).
This video shows a
robot observing a movie using common sense. The
flashing text and freeze frames are the internal
thoughts of the robot and not instruction text for
the viewers. These internal thoughts describe the
details of how the robot produce intelligence.
My robot doesn't use:
planning programs/heuristic searches (used by MIT and
Stanford University), Bayesian's probability theories
for decision making, Bayesian's equation for
induction and deduction, semantic networks for
natural language understanding, predicate calculus,
common sense systems, first-order logic, rule-based
systems, genetic programming, or MACHINE LEARNING.
When the robot
watches a movie (observation mode), a lot of
knowledge is generated on characters and scenes.
These knowledge include: identifying characters,
recalling facts about a character or groups of
characters, generating common sense knowledge,
producing logical facts, determining what body
movements or facial features imply, identifying inner
thoughts of characters, predicting the future of what
might happen in the movie, keeping track of the plot,
generalizing plots and complex scenes, and so on.
These extra knowledge
generated on the movie gives the robot addition data
about the movie. He is able to better understand what
is really going on and to see the hidden plots behind
the movie. Also, background information about each
character is vital to understanding the plot.
In the video, the
robot is observing blurry images and movie clips.
Because the robot is intelligent at a human level, he
can use logic and common sense to identify what
objects are in blurry images or movie clips. Let's
say the robot needs contacts and one day he loses his
contacts, that doesn't make him shut down because he
is unable to understand the world around him,
visually. He is still able to adapt and understand
that this blurry object is this object. He basically
takes an object from what he sees from the
environment and he extracts the closest match from
memory. His brain compares the 2 images (fuzzy image
from environment and clear image from memory) and
determine if they are the same things.
Despite images or
movie clips being blurry or unrecognizable, the
robot's brain is still able to understand what that
object is. The most important thing is that the
robot's mind is able to identify objects, actions,
and things regardless of the quality media. The media
type can be blurry, pixelized, up-side down,
distorted, stretched or partly missing. The robot's
intelligent pathways are able to use common sense to
identify the contents in any media type.
In other words, the
intelligent pathways in the robot's brain can form
computer programs in the robot's mind, to process
images. No human programmers are needed to write
codes to compare blurry images to clear images or
identify if an image is up-side down. The robot's
mind has a "self-defining image processor" to process
any image function.