= 27d567bc2d48d9f40e9ccc20ea370b15)
gjm comments on Generalizing From One Example - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (386)
Every time I have heard the phrase "mind projection fallacy" before, it has been with an entirely different meaning, namely the error of mistaking bits of your mental processes for aspects of the external world. It's unfortunate that it sounds so similar both to "typical mind fallacy" and "projection".
And a better name for the Mind Projection Fallacy is "Stealth Computation".
Why is that a better name?
If nothing else, its definition is more likely to be remembered separately from "projection" and "typical mind fallacy".
Well, sure, but on the other hand it's more likely to be thought of as (e.g.) a term for unconscious brain activity, or for thinking people do that isn't apparent to others, or for any phenomenon in the natural world that has computational power despite not having an obvious computing mechanism (e.g., evolution). And, at least to my mind, it has no particular connection with the phenomenon it's supposed to name. What I'm not seeing is why "stealth computation" is, overall, a better name than "mind projection fallacy".
I apologize for the diversion but would be most interested to hear your reasoning behind the attribution of computational power to evolution . (I presume you are referring to the process of evolution of living systems by natural selection) PK
I'd guess it goes something like this: the answer being computed is what set of genes is best adapted to the environment (a search problem over the space of reachable organism genomes); each organism is a possible answer; every generation, an organism producing more or fewer than the average # of offspring represents a computed 1 or 0; after enough generations... Not a Universal Turing Machine, no, but still computation.
Eliezer gives a few examples of this kind of thinking in http://www.scribd.com/doc/2327578/Worlds-Most-Important-Math-Problem-Eliezer-Yudkowsky-Future-Salon and I gather it's a reasonably well-established way of mathematically approaching evolution.
Yes, what gwern said. Evolution produces (very slowly and wastefully) things that are well adapted to their environments. It seems reasonable to call this an instance of computational power. If you (PK) prefer not to, though, fair enough; I think we would only be disagreeing about words, not about things.
OK you're right.