time is a coordinate
you’re taking the word “coordinates” too literally
also I’m not engaging any further, you’re pissing me off
programjames
Karma: 472
“eventually lead to a BB” is not the same as “locate a BB”
you do not have a BB unless you can point to it
I don’t think you should penalize Boltzmann brains or Earths (or sapient species) based on which particular one you are looking at.
But presumably you could also specify a recipe that eventually produces a BB.
I think the point of it being a Boltzmann brain is that there is no recipe. It’s just a random fluctuation. If there were a recipe, there would be time-consistency and it wouldn’t be a Boltzmann brain.
I think we’re looking at this from two different directions. Mine is, once trained, how many bits are in the structure of the brain? Yours is, during training, how many bits does the brain absorb? The numbers should come out the same, but mine is easier to get a lower bound and yours is easier to get an upper bound with. You have to worry about issues like, “what about memory that isn’t being accessed in this compute cycle?” while I made the assumption that every parameter is contributing 1 bit of information to the active computation.
My assumption is actually too strong though. Mixture of experts poses a problem, but a bigger problem is that it actually takes many parameters per bit of information. Probably. I mean, you only need to add a few hundred bits to train MNIST to >95% accuracy, while it takes ~10k CNN parameters.
(note: this is trained on a 300k ResNet, not a CNN).
I agree, a one-trillion parameter model (~4Tb) can probably simulate the brain well enough, and you might be able to squeeze out another order or two of magnitude. The brain only has a hundred billion neurons after all, and maybe the reason it needs many more synapses is because its clock speed is a hundred million times slower than a GPU.
ETA: Also, only ~1% of human genes are coding, so you arguably only need 10 Mb for the evolved brain.
I know people dealing with Long COVID. The root cause, the COVID virus, is easily identifiable and outside their body. The only way this helps with treatment is knowing they’re not dying from cancer. What does help? Figuring out all the inside-the-body issues, like post-nasal drip, vagus nerve dysfunction, gut issues, vitamin imbalance, blood clotting, immune suppression, and then going and treating those.
Lots of people break arms. Sometimes it’s a lifestyle issue. Maybe they should go mountain biking less often. Even if they threw away their bike, it wouldn’t fix the currently-broken arm.
Many times—probably most times—damage is closer to random than recurring, and looking for causes outside the body has marginal gain compared to just fixing the body.
If all mathematical structures exist, why do comparatively tiny numbers of orderly observer-moments seem to carry so much more weight-of-existence than the vastly more numerous horde of possible disorderly moments of conscious awareness?
Solomonoff induction. How many bits does it take to describe an disorderly moment of conscious experience? There are >100 trillion synapses in the human brain; let’s say around 1 Pb. Compare that to the bits needed to:
Start with a few axioms
Identify a planet with self-replicators
Describe the path of evolution to orderly brains
The vast majority of the bits here come from describing the DNA. Even if you do not compress it, with evolutionary path or genozip, this is less than 1 Gb. Perhaps in a fixed-size universe there will be an infinite number of Boltzmann brains, but locating any of them takes at least 1 Pb.
The difference between maths and non-maths is, in fact, mathematical. “That field over there? If you look at its structure, it does not use a hard logic where facts follow from axioms and tautologies. In fact, if you want to rigorously determine how strictly it adheres to that kind of structure… let me pull out my notepad and figure out how that should be defined.”
The SAT or ACT needs to be a hard legal requirement for all college applications everywhere
I could get on board with some examination being a legal requirement, at least for publicly funded schools (pretty much all of them), but the SAT/ACT have historically been very bad at maintaining rigor in an effort to increase profits. Just to pull a couple graphs from Wikipedia:
This is why a good SAT/ACT score is no longer enough for admissions at top schools.
wannanada looks like わんななだ rather than わんあなだ, not sure how to distinguish these in romaji. also tuwardu should be tuwaadu or tuwarudu.
I’ve found myself giving explanations (not as exonerations) when I suspect the other person is looking for a solution to their problem but does not know the levers to pull.
It does not matter if there is something outside maths if it does not effect you.
Not sure if that last sentence is sarcastic, but exactly! It is very problematic to babble (ctrl+f for “babbling”).
Thank you for sharing these resources. I saw you talking about several nonobvious things in your post (field theory, morphisms weighted by Kolmogorov complexity), but was very thrown off by your use of “phenomenal” and “qualia”. Usually I just strong down vote such posts and move on, but given the rest of your post decided to query for more information. I could have been nicer in asking, but I don’t think infectious diseases deserve to be treated nicely. They should be quarantined and disinfected. (Describing the terms “phenomenological” and “qualia” here.)
The issue with these terms is they were created in opposition to consciousness. As in, “no, a simulated/artificial brain is not actually conscious, it doesn’t have
phlogistonphenomenal consciousness”. For a similar reason, I do not like the term “access consciousness”. There is just consciousness, and being finite beings, what we can access of it.I skimmed through Max Hodak’s talk and it matches my intuition. I think our ideas of consciousness are mostly the same, including the field theory of memes.
Based on this, I can translate what you are saying by making the word maps:
“qualia” → “particle/meme”
“phenomenal”→”″
I can understand your adoption of religious language (even though you are not referencing the same thing as the believers) to avoid being labelled a heretic, especially because this is your research area and you do not want to lose funding.
I read a couple of your blog posts, and they are interesting, mostly because of your math background. I’ll probably read more later. Again, thanks for responding despite my pessimism.
This is an equivalent question that may help you understand my frustration better. What extensional properties of “phenomenal consciousness” are there that distinguish it from “access consciousness”?
I did read the post before commenting. Obviously. And yeah, I did see that “equivalent to a symmetry group” thing going on. Less obviously.
Your post is just one of many where I see the terms “qualia” and “phenomenal consciousness” thrown around without any attempt to define them. Every time I look into it, I see a mysterious answer to a question no one was asking.
At this point, I’ve decided these words are simply nonsense, just stupid memes floating around no different than saying, “humans are special because they have souls!” Given you spend a lot of time thinking about this, do you have a constructive definition of these terms you can point me to? A definition that does not end in, “you must use your blind faith,” or, “it’s a axiom you functard.”
The crux of our disagreement shows up in this little metagame. You think people need to be told what to think, while I think they should think for themselves. Your thinking might fail, but you will be better off for trying. At the very least, you will not frustrate your interlocutor asking for verbose explanations, because they can show you one corner of a square and you can find the other three.
My first reply was demonstrating how thinking for yourself does work. You do not need literature, just a steady stream of problems to forge your critical thinking skills against.
From your reply,
I agree in many cases this ability is bounded by an individual’s intrinsic talent at doing so—whether by desire, intellect, hardiness, or such.
I realized you also think it can work, but only for those privileged enough with intrinsic talent. From your original comment, the argument seems to be: critical thinking can practiced and is useful, but only for those privileged enough to spend time on the endeavor; most people are better off just learning from books.
I strongly disagree with this idea that critical thinking is for the privileged. I think it is quite the opposite: only the privileged grow up in cultures that show them the best resources to learn from. Everyone else gets indoctrinated into the local cult, where football or magic worship takes precedence. Even those a little luckier learn studying is important, but being the students have no way to differentiate between “Linear Algebra Done Wrong”, “Linear Algebra Done Right”, Khan Academy, and MIT OCW, assuming they even find these resources.
Every state usually has one or two MATHCOUNTS clubs that win almost every year. In my day, it was Frost Math in Massachusetts, Quail Valley in Texas, and Desert Ridge in New Mexico. Why, even in very competitive states, is this so? Because some are privileged to be taught by better coaches, with better resources.
Yes, with the right resources, you can learn a lot faster from emulation than innovation. Only the very privileged get these resources. Everyone else is better off developing their critical thinking skills. How can you call it a privilege to think for yourself, when very few seeking to rise up the ranks get the opportunity to not think for themselves?
Why didn’t you already recruit a model? Take a screenshot and feed it into Claude. I just tried, it worked. So what was the need for these two comments?
I think it would be more interesting to play deception chess with another human or an AI trained adversarially against humans in particular (such as an easier version of LeelaQueenOdds). The issue with most bots is they do not setup traps. The deceivers have to somehow get their advisee to actively make mistakes, rather than passively not see a trap.