Growing on to Othello

During the last few months, I've been developing strong Othello playing systems which is my major area of research at the present. Initially I was kind of relunctant to research in this area, as I was far more interested in other topics, but due to unforseen circumstances, I was kind of forced to explore in this domain.

Anyway, though at times you don't appreciate what you are forced to do, I have to admit, once you start to understand the naunces and challenges of the new task, you start to appreciate it more - and at times I have to admit, you start to love it.

All I want to say at this moment is that I've been working with two approaches at the present. The first is using a Genetic Approach to evolve board evaluation functions for Othello. In this case, I've adapted Charles Darwin's theory of evolution to come up with board evaluation for my Othello system. This approach though seems promising at first, I have to admit after doing numerous experiments with the numerous variations of the Evolutionary approach in programming, it ended up not as good as anticipated. To put it in CS terms, I feel that the Genetic Approach is just an heuristic for a brute force search on certain problem domains. I was looking into using parallel processing to help speed up the evolution and hope that the performance would be better - but after consideration, I decided that this approach may not be the best way forward.

After a few weeks of thinking, I've come up with a new radical idea that is so simple, yet very effective. The idea is simple, yet profound to a point I have a reason to believe that this system has potential to actually beat many of the World Class Othello playing algorithms once it is done. At the moment, I'm finishing up the prototype system and initial experiments prove to be extrememly promising. Once the system is up and running, I'll explain more on this idea.

Here I come, Pisal the Othello Expert ;)