Analyzing predator 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).
In the video, a robot is
watching the movie, predator, and answers questions
based on what he saw.
In early 2000s, I
went to my college professor and asked him one
question: are there any robots built that can answer
simple questions from kids books? My professor
explained to me that it is very difficult to build a
robot that can read and answer questions like a 5
year old. He also stated that such technology don't
exist yet. He used popular Artificial Intelligence
books at that time to back up his statements.
This fact is also
backed up by Darpa because in 2012 Darpa's robot
challenge was announced, which is a contest to build
a robot that can think and act like a 5 year old.
They want to build a robot that can climb ladders,
navigate around obstacles and use basic tools to do
things.
This conversation
with my professor told me one thing: no one has
designed a robot that can think and act like a 5 year
old yet. My goal was to design a robot that can think
and act like a human adult, capable of human level
intelligence. In order to do that I had to analyze my
own brain and find out how it thinks while doing
human tasks. I would document every data that comes
through my mind and put that in a notebook. I would
do this for driving a car, cooking food, answering
questions, playing videogames, or cleaning my house.
Answering questions
from a movie is one (out of many) skills a human
being has. Human beings can do any human task and all
knowledge is learned from teachers in school.
When the robot
watches a movie, thoughts activate and these thoughts
include: comments, analytical data, observation,
identify objects, comparing objects, logical facts,
personal opinion, likes/dislikes, etc. The robot
learned to analyze a video or picture from teachers
in school. These analytical thoughts from the robot
are stored within the movie. These robot thoughts
help to search for data and it helps to organize data
in the movie.
In the video, one of
the questions asked was: how come the predator didn't
find the man? The answering of this question is
really complex because the robot needs to know a lot
about common sense and object properties. The robot
needs to know that when someone touches mud, the mud
sticks to you. He needs to know that when mud covers
the entire body, no body heat is emitted. Next, the
robot needs to know that a reptile sees the world
through heat signatures. If the man was covered in
mud, he doesn't emit heat, the result is the predator
(which is a reptile) can't see the man. Thus, the
answer to the question is: the predator can't see the
man because he was covered in mud.
In the video, the
second question asked was: what is the color of the
laser? In the movie there are no facts given about
what the color of the laser is. The robot has to
search for pictures of the predator using the laser
and id what the color is. by the way, the knowledge
to id the color of an object is based on lessons from
teachers. Teachers teach the robot how to id colors
from objects. The robot is simply using this
knowledge to extract the color of the laser from a
picture in memory.