donderdag 6 december 2012

Quote #7: Amy Perfors

One implication of this is that Fodor's point is true but trivial. If one understands ‘‘concept’’ to mean ‘‘something that can be represented by the brain’’, then all of our concepts are innate – they exist in the latent hypothesis space of possible things the brain can represent (a space implicitly defined by the structure of the brain). In the more interesting sense, where ‘‘having concept’’ means that the concept is available at the cognitive level – it is capable of being manipulated by the cognitive system – then it need not be innate (i.e., having always been available at that level).

Amy Perfors (2012). Bayesian Models of Cognition. What's Built In After All? Philosophy Compass 7/2, p. 132

dinsdag 20 november 2012

Arab Strap - The Shy Retirer

Quote #6: Edward Tsang (2008)

It is also worth noting that computation itself involves a cost. Knowledge acquisition (e.g. to find out the travelling costs between two cities) could also involve costs. A rational agent should not only minimize travelling cost. It should attempt to minimize the travelling cost plus the cost of computation and knowledge acquisition. (p. 63-64)

Tsang, E. P. K. (2008). Computational intelligence determines effective rationality. International Journal of Automation and Computing 5(1), 63–66.