| Forum Index > General Forum |
|
farvacola
United States18818 Posts
On July 08 2013 04:43 Rhaegal wrote: Of course it is. What else could it be? We're no more special than a rock. Though you have to admit, rocks have a poor time obsessing about their own death in poetic form. | ||
|
Shiori
3815 Posts
On July 08 2013 04:43 Signet wrote: I've only read a small amount of his essays. My general understanding is that the scientific method and Bayesian hypothesis-updating are almost always compatible, and that the scientific method generally gives you very very strong Bayesian evidence. The kind that would cause a Bayesian rationalist to change his beliefs unless he had very very strong priors. This essay: http://lesswrong.com/lw/qa/the_dilemma_science_or_bayes/ seems to come down to a very specific case -- what happens when two theories are equally supported by evidence, and one theory was proposed first but the other theory is "simpler" (from an information-theoretic perspective)? The scientific method says the first theory remains accepted until proven otherwise, Bayes says the second. That's incredibly rare. The example EY gives - wavefunction collapse vs Everett many-worlds interpretation -- is probably one of the few that even exists. This doesn't seem to me like rejecting the scientific method. More like saying that it works 99.999% of the time (as does Bayes which comes to the same conclusions), and there's this small fraction of the time when it might make more sense to use Bayes instead. But yeah I'm not really a LWer, just have read some of the essays... has he said anything more strongly against the scientific method than what I linked? I'm mostly just curious for curiosity's sake, thanks  It's definitely rare, but the example he uses is actually a really big deal. The MWI model is not something the majority of physicists are in favour of with any certainty; throughout the entire field, it would be wrong to state that most serious quantum physicists hold the MWI to be head and shoulders above other interpretations. I mean, the essential problem with EY is that he uses Bayesian updating to shimmy in really controversial conclusions (like MWI) without making it seem like a big deal. This is a piece I agree with regarding the article in question. | ||
|
Roe
Canada6002 Posts
On July 08 2013 04:49 xM(Z wrote: that is false. what you meant to say was: the theory of evolution has been corrected again and again and again. someone who i can't remember said it better than me: there's only a dispute about how we evolve, not that we evolve. | ||
|
wherebugsgo
Japan10647 Posts
On July 08 2013 04:31 politik wrote: I was replying in context to your statement of "determination and free will are not mutually exclusive" And they're not. Determinism deals with cause and effect-that, for everything that happens, given a set of conditions, nothing else could happen. The compatibility comes in where free will has to do with a choice between events-between "effects" basically. Determinism only really applies to events that have already happened. Free will applies to events that have not yet happened. Pretty stark difference and why they are indeed compatible. After a choice has been made, certain conditions have been satisfied. Events leading up to the choice are deterministic because you can observe what has already happened. The process that governs the choice need not be deterministic. This is pretty apparent when you realize that our current best models of the universe (based on quantum mechanics) imply that the act of observation affects our results. Consider this thought experiment: I decide to flip a coin-if heads, I will do X. If tails, I will not do X. The events leading up to the decision to flip the coin can be mapped as a chain of causes and effects. The result of the choice however is not deterministic until AFTER it happens. Thus, it is only in hindsight that the conditions are met. You don't know if future conditions will be met, which is why your assumption (that I am guaranteed to accept something after reading it-I have not read it yet) is bad. No, I have a choice because the events that happen after the present are not predictable-they are not purely representable deterministically until they have already happened, and my choice depends on the knowledge of events that have already happened. When my choice is made, it is deterministic-meaning, given the conditions that lead to me making the choice (at the instant that I did) it's impossible to do anything differently. Again, this conclusion is worthless because it has no predictive power-you already know everything that happened. It's not worthless based on any emotion, but based on the fact that it really doesn't tell you anything or even suggest anything you already didn't know. | ||
|
DoubleReed
United States4130 Posts
On July 08 2013 04:53 Shiori wrote: It's definitely rare, but the example he uses is actually a really big deal. The MWI model is not something the majority of physicists are in favour of with any certainty; throughout the entire field, it would be wrong to state that most serious quantum physicists hold the MWI to be head and shoulders above other interpretations. I mean, the essential problem with EY is that he uses Bayesian updating to shimmy in really controversial conclusions (like MWI) without making it seem like a big deal. This is a piece I agree with regarding the article in question. So I got the impression that your hate for Eliezer Yudowsky was a bit more passionate than a simple disagreement on quantum mechanics, or even some applications of Bayesian Reasoning. I'm not really into the transhumanism or singularity stuff myself. That's usually where he goes into more kooky territory. | ||
|
oneofthem
Cayman Islands24199 Posts
On July 08 2013 04:25 Snusmumriken wrote: you are cutting eepiphenomenalism too little slack. if u take it to use a negative definition of qualia as whatever that is not physical then kim's distinction begins to make some sense. what he says is that what you feel is meh, but the structural info mirrored by the feeling is legit. will have to get myself to a computer for this...First of all this is only true if phenomenal consciousness is an epiphenomenon, and epiphenomenalism is a metaphysical dead-end in my opinion (and not all epiphenomenalists would accept the possibility of zombies). Second, its only true in a trivial sense. It is similar to the old beetle in the box problem by wittgenstein. The zombie would say he has a beetle in the box even if there is no beetle in the box, but the fact remains: In the zombies case, there IS no beetle in the box. Similarly, the zombie would say he is conscious, but he would not be. If we exchange the zombie for a computer who has beliefs in the same sense a zombie would have beliefs, ie not at all in what we normally call conscious beliefs as they have myriads of phenomenal qualities attatched to them, then what youre saying is more or less that a computer programmed to answer "yes" when asked "are you conscious" somehow shows how we cannot doubt our own consciousness "even if its not there". The whole problem arises because language is sloppy and "belief" has no strict definition. If it had we would have to use different words for "zombie-belief" and our own belief as the zombie-belief is no more than how a computer "beliefs". The computer does not feel pain, even if it "believes" it does. To me there is no issue of not being able to rule out being a zombie, because the one thing that is self-evident to me (and presumably all self-conscious beings) is that im conscious. If I truly was a zombie then I would still "believe" that in a sense, but only in a trivial sense and as a result of the vagueness of natural language. but imo the problem is not so much vagueness of natural language (or that answer is itself too vague ). it's really about trying to represent mental representation while being restricted to incommensurate representational faculties . we can seemingly identify the same situation mentally (pain) as well as in terms of physical description (neuron firing). the mental object populated description doesn't represent its own physical causal mechanism (intuit epiphenomenalism ), but is itself factually identical to pure physical scenario (reductive physicalism ), yet does represent some thing in the world that is not commensurable with physical representations of the same situation . basically the qualia populated story is about le world, but the way it latches, identifies with the world is through the sense of self(or sense of bat lol. indexicals) such stories engage. in a heisenberg twist of sort the physical account gives the facts but cannot give the experience and for the qualia story vice versa. but qualia stories always piggy back on a subject indexical (whose identity is trivial precisely because of the indexical nature) for its hook onto the world. taken by itself it becomes a dubious abstract object with no location the abstract object with a world ly shadow , but we ought to understand that it is still only a shadow | ||
|
Shiori
3815 Posts
On July 08 2013 05:11 DoubleReed wrote: So I got the impression that your hate for Eliezer Yudowsky was a bit more passionate than a simple disagreement on quantum mechanics, or even some applications of Bayesian Reasoning. I'm not really into the transhumanism or singularity stuff myself. That's usually where he goes into more kooky territory. I like some of his work on eliminating cognitive biases (like well-established psychological ones rather than obscure/discovered ones). I think that's basically the only area in which he says unequivocally useful and (mostly) true things. When he talks about ethics viz. idealized AI, transhumanism, or anything about singularities, I find him very depressing/crazy/strange. My hate for him derives solely from the pseudo cult of personality which has developed around him, in which he's something like a ubermensch and grand arbiter of truth/rationality. Also it hurts me that his Harry Potter fanfiction is so adored by some people because I think it's really awful as a piece of literature. (I don't actually hate him. I just can't stand his worldview and its emphasis on futurism and analogizing everything to idealized AIs). I also find that it's nearly impossible to debate with his acolytes because they tend to have this entire system worked out. It reminds me of debating with Austrian school economists/libertarians. It's like: you just want to debate the issue at hand, and yet they keep introducing more concepts from their overarching system and you're just like ughhhh. | ||
|
Signet
United States1718 Posts
On July 08 2013 04:53 Shiori wrote: It's definitely rare, but the example he uses is actually a really big deal. The MWI model is not something the majority of physicists are in favour of with any certainty; throughout the entire field, it would be wrong to state that most serious quantum physicists hold the MWI to be head and shoulders above other interpretations. I mean, the essential problem with EY is that he uses Bayesian updating to shimmy in really controversial conclusions (like MWI) without making it seem like a big deal. This is a piece I agree with regarding the article in question. Yeah, I can definitely see that in his writing... and as DoubleReed says, even more so with some of the stuff about singularity, simulated consciousness, etc. With the rebuttal article on MWI, it seems like Eliezer perhaps didn't make it clear that his idealized form of Bayesian reasoning comes from Solomonoff induction... within that particular framework, it follows that, since (algorithmically) MWI is strictly simpler than wavefunction collapse, the fact that the evidence supports both equally well means that MWI should be seen with a higher probability of being correct than Copenhagen. Not a probability of 1, but something proportional with however many bits of information it would take to specify a program simulating a universe with MWI versus a program simulating a universe with wavefunction collapse. Of course that is not universally accepted as the ideal form of Bayesian Reasoning. And MWI is not universally or even overwhelmingly accepted among quantum physicists. So that does explain your leeriness of things sounding too lesswrongy. | ||
|
DertoQq
France906 Posts
On July 08 2013 04:51 farvacola wrote: Though you have to admit, rocks have a poor time obsessing about their own death in poetic form. I don't know, I never had the chance to talk with one of them. | ||
|
Signet
United States1718 Posts
On July 08 2013 05:30 DertoQq wrote: I don't know, I never had the chance to talk with one of them. For all we know, rocks could have the same epiphenomenal experiences that the rest of us do! They just don't have the p-zombie outer shell which goes on rambling about it. | ||
|
YumYumGranola
Canada344 Posts
| ||
|
Shiori
3815 Posts
On July 08 2013 05:39 YumYumGranola wrote: I feel like there's an awful lot of mental masturbation going on in this thread. Actually it's just chemicals and electricity tangoing with the same. | ||
|
AzureD
United States320 Posts
On July 08 2013 04:55 Roe wrote: someone who i can't remember said it better than me: there's only a dispute about how we evolve, not that we evolve. Even then the how is in the small details. What we do not know is the path of evolution that most of life has taken because fossils are rare so an unbroken chain of life is impossible to find. Like a movie that is missing most of its frames. It is difficult if not often impossible to find a direct ancestor. Most of the time we find cousins of the direct ancestors. Fossils are only supporting evidence for evolution. It was not through fossils that evolution was determined to be most likely correct. | ||
|
YumYumGranola
Canada344 Posts
On July 08 2013 05:45 AzureD wrote: Even then the how is in the small details. What we do not know is the path of evolution that most of life has taken because fossils are rare so an unbroken chain of life is impossible to find. Like a movie that is missing most of its frames. It is difficult if not often impossible to find a direct ancestor. Most of the time we find cousins of the direct ancestors. Fossils are only supporting evidence for evolution. It was not through fossils that evolution was determined to be most likely correct. I think you're underestimating the quantity and quality of fossils that we've now gathered. There are of course other corroborating methods such as analyzing genealogy for evidence of common descent. | ||
|
oneofthem
Cayman Islands24199 Posts
| ||
|
wherebugsgo
Japan10647 Posts
On July 08 2013 05:33 Signet wrote: For all we know, rocks could have the same epiphenomenal experiences that the rest of us do! They just don't have the p-zombie outer shell which goes on rambling about it. For the sake of your intelligence/value of opinion I hope this statement was satirical or sarcastic in nature. edit: It kinda reminds me a lot of a Hindu guy I know who always tries to reason with me that it's very possible that plants have feelings just like humans-even that they can feel pain. It's really ridiculous, even coming from him, given that he is a physician. | ||
|
Signet
United States1718 Posts
| ||
|
Shiori
3815 Posts
On July 08 2013 06:45 Signet wrote: Satire. Although I will note that the idea of an epiphenomenal consciousness that lacks a body capable of expressing that consciousness isn't any more ridiculous than the idea of philosophical zombies writing books about their own consciousness that they are not actually experiencing. Philosophical zombies are thought experiments. Obviously they're ridiculous. That's the point. It's kinda like Schrodinger's Cat. Obviously the example itself is utterly silly, but that's the point of the example: to be silly and yet still obey the parameters of some stated system. | ||
|
oneofthem
Cayman Islands24199 Posts
| ||
|
Signet
United States1718 Posts
On July 08 2013 06:52 Shiori wrote: Philosophical zombies are thought experiments. Obviously they're ridiculous. That's the point. It's kinda like Schrodinger's Cat. Obviously the example itself is utterly silly, but that's the point of the example: to be silly and yet still obey the parameters of some stated system. Yeah I know Look at the rock that doesn't show any evidence of consciousness but nonetheless experiences it epiphenomenally in the same way. If consciousness were truly epiphenomenal (which is an actual philosophical position), there's no reason it couldn't exist outside of a human brain. | ||
| ||
StarCraft 2 StarCraft: Brood War Dota 2 League of Legends Counter-Strike Super Smash Bros Heroes of the Storm Other Games Grubby11059 FrodaN2605 Beastyqt1115 shahzam449 elazer343 Pyrionflax230 C9.Mang0139 ZombieGrub62 JuggernautJason52 minikerr48 Railgan23 OptimusSC213 |
|
Tenacious Turtle Tussle
PiG Sty Festival
Serral vs TriGGeR
Cure vs SHIN
The PondCast
Replay Cast
PiG Sty Festival
Clem vs Bunny
Solar vs Zoun
Replay Cast
Korean StarCraft League
PiG Sty Festival
herO vs Rogue
ByuN vs SKillous
SC Evo Complete
[BSL 2025] Weekly
[ Show More ] |
|
|
EPT
