(Ultimate smoking vacant chaos version.)
Might groggily try to poke some bits and pieces in here later …
On copying. Correction.
[…] By admin […]
[…] Source: Outside In […]
What is the Yudkowsky/LW model of intelligence?
I imagine the universe would die a heat death before random brute force tinkering would yield anything. I think it’s more to do with a combination of teaching, copying, transcribing, and innate intelligence. Homo sapiens have an intrinsic intelligence due to the neurological wiring of into the brain structure that allows them to tinker, deduce/infer, and communicate such results. A less intelligent person may only be able to copy; a more intelligent one can make inferences.
I thought you’d include the ultimate escape of the year in this patch.
About copying. Replicator teleology is *the* core NRx idea: “the only morality is civilization”.
Exfernal Reply:January 5th, 2016 at 1:18 pm
“Descent with modification” is a rather neat expression.
The piece on copying is very confused. Reproduction, sex, and copying are not the same thing. Meiosis and mitosis are not the same thing.
It’s a sloppy rendering of Aristotle’s metaphysics whereby the author substitutes “copying” for Aristotle’s four causes and then posits that in biology the cause is teleological, the final cause.
Teleology in biology and Darwinism is an age old controversy. It’s not a new idea or insight. Also various conservatives, libertarians, etc. have been trying to revive Aristotelian virtue ethics for a long time now, and it’s incorporated into Catholic dogma.
Aristotelian teleology does not make good predictions. It makes good ad hoc explanations that seem to make sense but vary wildly. Consider Aristotle’s own telos for man, happiness, which differs significantly from the telos posited in the article.
Part of the reason they differ significantly is that contemporary attempts to revive Aristotelian virtue ethics tend to be post hoc rationalizations of emotions and preferences.
Alrenous Reply:December 28th, 2015 at 3:35 pm
Can rescue teleology by identifying it with equilibrium. The final equilibrium of democracy, for example, is chaos and destruction.
However, biology has no final state. It could conceivably go on changing and improving forever. With a given environment there may be some small collection of stable final states, but the environment is partly made of biology, meaning as evolution continues, the environment changes.
Splice in the fact that the perfectible is less than the imperfectable. By definition, the perfect has some finite best state. The imperfectable can go on improving without limit.
NRx_N00B Reply:December 28th, 2015 at 6:31 pm
I guess biology acts more/less plastic and tries to equilibrate/optimize itself to the current environment. If the environment changes too rapidly it goes from being ductile to brittle, then shatters and goes extinct. ‘Perfectibility’ is a relative thing that is in a constant state of flux—it morphs with the changing environment.
Erik Reply:December 29th, 2015 at 12:21 am
“Splice in the fact that the perfectible is less than the imperfectable. By definition, the perfect has some finite best state. The imperfectable can go on improving without limit.”
I am unconvinced of this. The imperfectable might for instance be undergoing continual improvement in the face of recurring damage – toy example: t minus floor(t). Or the imperfectable might be imperfectable because it is stuck in a non-best final stable attractor state inferior to the perfect state. The argument could be rescued by saying that (some) imperfectable undergoes strict monotone improvement indefinitely, but this does not seem obviously true.
SVErshov Reply:December 29th, 2015 at 2:34 am
example of “perfect” can be hydra, practically eternal organism, if environment is good.
NRx_N00B Reply:December 29th, 2015 at 4:48 am
“example of “perfect” can be hydra, practically eternal organism, if environment is good.”
As far as the gene-centered view of evolution goes:
In stable environments, immortality might be a good strategy, but in environments that are subject to change, mortality is what gives life the plasticity needed to adjust to a moving/drifting steady-state (i.e. “the ends” is a moving target—”improvement” doesn’t shoot at a fixed target).
Empedocles Reply:December 28th, 2015 at 5:47 pm
“Meiosis and mitosis are not the same thing. ” But they are both the copying of genes.
For virtue ethics, see here: http://darwinianreactionary.wordpress.com/2015/08/26/restoring-a-virtue-based-ethics-for-the-21st-century/
Bob Reply:December 29th, 2015 at 12:09 am
In mitosis, genes are simply copied. In meiosis, genes aren’t simply copied but discarded and combined with other genes.
What’s the difference between a correction and a collapse? I guess a collapse is a monster correction—a full margin rupture—that happens when a system is too rigid and isn’t allowed to adjust itself with minor corrections to relieve the build-up of stress (i.e. snapping back and ‘undershooting’ the steady-state from a position of ‘overshoot’); translatable into pretty much anything that can be described via systems dynamics.
Grotesque Body Reply:December 28th, 2015 at 10:18 pm
I think of correction as modification of the activity/policy/tactics of a system, while preserving its axiomatic identity/objectives, while collapse undermines its axioms completely.
Correction happens to antifragile things, while collapse happens to fragile things. A system undergoes correction through collapse of some but not all of its subcomponents.
NRx_N00B Reply:December 28th, 2015 at 11:39 pm
Yeah, that’s another good way of looking at it. Once a system crosses a certain threshold (i.e. distance from the steady-state) it goes from correctable/’antifragile’ to crashable/’fragile’. Somewhat analogous to a solid-state polymorphic phase transition, where the system is undermined and becomes something entirely new/stable (i.e. rapid, almost instantaneous, re-equilibration).
NRx_N00B Reply:December 29th, 2015 at 12:39 am
In political terms:
Trumpenführer will keep the system beneath the threshold = correction != popcorn worthy
Hillary will push the system beyond the threshold = collapse = Nemesis = popcorn worthy
Grotesque Body Reply:December 29th, 2015 at 12:53 am
However Cathedral reaction to Trump candidacy and Trump policies = popcorn worthy.
sportsbet.com.au currently suggests that Rubio is most likely to get the GOP nomination (although Trump is quickly closing the distance, down to 50 cents now) while Nemesis is overwhelmingly likely to get POTUS. It will be interested to see what happens to the odds for elected POTUS once Trump is tipped to get the nomination.
Somewhat analogous to a solid-state polymorphic phase transition, where the system is undermined and becomes something entirely new/stable (i.e. rapid, almost instantaneous, re-equilibration).
Quite a bit of effort went into studying chaotic systems back in the 1960s-1980s.
One thing they found is that equilibrium does not exist except as a temporary condition within a larger periodistic order.
That is, you cannot have equilibrium without some form of goal, akin to “teleology” which seems to be the new anti-trend.
NRx_N00B Reply:December 30th, 2015 at 5:19 am
Thanks Brett, I was being a little sloppy and using “steady-state” synonymously with “dynamic equilibrium” (or goal-seeking, where the goal is dynamic and in a constant state of flux). Since systems exist at nested hierarchical levels, your “periodistic order”, is yet another system but at a much larger scale (“periodistic” → an oscillating system?; not unlike predator-prey population cycles, business cycles, tidal cycles, the human heart, civilization cycles…etc).
As far as “teleology” goes, I think external-teleology (i.e. a goal that is the product of a mind, an agent, with intentions/purpose) is completely useless in biology or science at large. The only universal or overarching goal/tendency is for the ball to roll to the bottom of the hill; the ultimate equilibrium, heat-death—thats it.
“where the system is undermined and becomes something entirely new/stable”
More sloppiness on my part, I should have been explicit; “lattice system”, for example, transitioning from a hexagonal to cubic system.
NRx_N00B Reply:January 1st, 2016 at 5:27 am
“The only universal or overarching goal/tendency is for the ball to roll to the bottom of the hill”
Events between A and B are about chaos, kinetics and nonlinear dynamics—it is the eddies, the von Kármán vortex sheets/streets, the convection, and eventual laminar flow..etc, that make life/spontaneous order possible/inevitable.
The religion of progress does weird things to minds of even the brightest among us: https://soundcloud.com/samharrisorg/surviving-the-cosmos
How can one be so intelligent and yet so oblivious about the nature of intelligence? Maybe orthogonality is true after all.
Mail (will not be published) (required)
Notify me of follow-up comments by email.
Notify me of new posts by email.