Based partially on this thread:
http://www.internationalskeptics.com...wpost&t=285993
Occasionally we have threads on the ISF about the tendency for skeptics to have specific topics that we are less than rational about. One of my personal topics that fall within this description is my belief in the power of random numbers.
To summarise the belief: it's essentially that with a little bit of clever manipulation most problems can be solved with the application of sufficent randomness. Not only that but that this approach will be the most efficient approach as well.
This is specifically in the case of computational problems.
To give an example:
Myself and my colleagues use LegoMindstorms (programmable lego robots) to teach basic programming to our students. One of the tasks that we ask the students to perform is to navigate past a number of obstacles on a open area.
My ideal solution would be for the robot to move forward until it finds an obstacle, reverse a small amount, turn in a random direction by a random number of degrees and try moving forward again.
I believe this is superior to having a fixed direction for the turn and a fixed number of degrees.
I think this belief comes from my final year Computational Intelligence classes where I was introduced to genetic algorithms, particle swarm optimizers and other techniques that relied heavily on random number generation. I think I was captivated by the elegance of using random numbers and this might have turned into a belief.
My question is, is this an irrational belief, or merely a misguided one, and how could one test it? Also does anyone else have this belief?
http://www.internationalskeptics.com...wpost&t=285993
Occasionally we have threads on the ISF about the tendency for skeptics to have specific topics that we are less than rational about. One of my personal topics that fall within this description is my belief in the power of random numbers.
To summarise the belief: it's essentially that with a little bit of clever manipulation most problems can be solved with the application of sufficent randomness. Not only that but that this approach will be the most efficient approach as well.
This is specifically in the case of computational problems.
To give an example:
Myself and my colleagues use LegoMindstorms (programmable lego robots) to teach basic programming to our students. One of the tasks that we ask the students to perform is to navigate past a number of obstacles on a open area.
My ideal solution would be for the robot to move forward until it finds an obstacle, reverse a small amount, turn in a random direction by a random number of degrees and try moving forward again.
I believe this is superior to having a fixed direction for the turn and a fixed number of degrees.
I think this belief comes from my final year Computational Intelligence classes where I was introduced to genetic algorithms, particle swarm optimizers and other techniques that relied heavily on random number generation. I think I was captivated by the elegance of using random numbers and this might have turned into a belief.
My question is, is this an irrational belief, or merely a misguided one, and how could one test it? Also does anyone else have this belief?
via International Skeptics Forum http://ift.tt/1vhYTYo
Aucun commentaire:
Enregistrer un commentaire