Sunday, July 09, 2006
IWLCS 2006: In the heat of the discussion
We are having a discussion session about what, where, and when next? The question that sparked the fire: "Is reinforcement learning the enemy?".
Comments:
<< Home
Hallo, Xavier! :)
What about trying opinions without manichaeism essence? :D
I suppose the readers of this blog and all EC enthusiasts deserve arguments that can reach much more than a simple "black vs white" question. :)
It is a pity that persons in the "The Most" (according to some people) Evolutionary Computation conference produce that kind of question. I guess there are researchers, practioners, etc from whole world attending GECCO-2006 and would not be interesting to bring argumentations which can be harmful for the collective, that is, the Evolutionary Computation field and other AI methods.
Enemy? Tell me: Which insights could that type of question bring to the LCS field (or to the whole EC field)? Wouldn't be more interesting to say: "What can we learn with other methods, such as Reinforcement Learning?" ?
Persons frequently do not admit their limitations and do not know how to deal with their deceptions. That is another problem.
Anyway, would be very wise to researchers to not put in others the responsability of failure of their (the researchers) method.
Até Mais!
Marcelo
Post a Comment
What about trying opinions without manichaeism essence? :D
I suppose the readers of this blog and all EC enthusiasts deserve arguments that can reach much more than a simple "black vs white" question. :)
It is a pity that persons in the "The Most" (according to some people) Evolutionary Computation conference produce that kind of question. I guess there are researchers, practioners, etc from whole world attending GECCO-2006 and would not be interesting to bring argumentations which can be harmful for the collective, that is, the Evolutionary Computation field and other AI methods.
Enemy? Tell me: Which insights could that type of question bring to the LCS field (or to the whole EC field)? Wouldn't be more interesting to say: "What can we learn with other methods, such as Reinforcement Learning?" ?
Persons frequently do not admit their limitations and do not know how to deal with their deceptions. That is another problem.
Anyway, would be very wise to researchers to not put in others the responsability of failure of their (the researchers) method.
Até Mais!
Marcelo
<< Home
