Transcript my_poster_2

Examining the Mechanisms of the Human Brain with Computer
Science
Arvind Ravichandran
2005-2006
Computer Systems Research
Period 5
Background:
The object of all artificial intelligence studies is to simulate the capabilities of the Human brain. No
project thus far has even been mildly successful in this endeavor. However, in this project, I seek to
instead use artificial intelligence to study the human brain, rather than vice versa. By studying current
methods of Artificial Intelligence and workings of the Human Brain, i identified one key possibility:
pattern recognition. The goal of this project was to simulate pattern recognition and see if it could be
used as a primary learning mechanism in the brain.
Num ber of Ties (as a Percent age)
Result s of t he Tic Tac Toe Gam e
0.2
0.175
0.15
0.125
0.1
0.075
0.05
0.025
0
1
2
3
4
5
6
7
8
9
Trial
Description:
In order to effectively complete this project, I did a number of things. First, I research current
methods of Artificial Intelligence, including those found in video games, robotics, and
mathematics departments. Then, I researched current research in the field if Human Learning
and neurology. I combined my findings from these two research endeavors to formulate my final
project idea: simulating pattern recognition to judge its effectiveness as a possible learning
agent. To construct the simulation, I created a program that could simulate as much of the
human brain as was possible in the time frame. This program included a message processing
center, data conversion, and crude memory storage. I used a process of layers of abstraction in
my program, which I found to be extremely useful and was a useful side result of my research.
Second, I programmed a tic tac toe game with an unbeatable artificial intelligence so that the
computer could practice. Finally, i ran the program repeatedly to obtain adequate results for
scientific analysis. The final results of my program indicated that pattern recognition alone would
not suffice as a learning mechanism. It was clear that the Human Brain did need an original
sense of correctness in order to interpret the results.