Human brains and experiences seem to be constrained by the laws of quantum physics, which can be simulated to arbitrary fidelity on a computer. Nit sure where Godel’s incompleteness theory would even come in here…
how are we going to deduce/measure/know the initialization and rules for consciousness? do you see any systems as not encodable/simulatable by quantum?
I think you are asking whether consciousness might be a fundamentally different “thing” from physics and thus hard or impossible to simulate.
I think there is abundant evidence that the answer is ‘no’. The main reason is that consciousness doesn’t give you new physics, it follows the same rules and restrictions. It seems to be “part of” the standard natural universe, not something distinct.
Can you look at any arbitrary program and tell if it halts without running it indefinitely? If so, you should explain how and collect your Nobel. Telling everybody whether the Collatz conjecture is correct is a good warm up. If not, you can’t solve the halting program either. What does that have to do with consciousness though?
Having read “I Am a Strange Loop” I do not believe Hofstadter indicates that the existence of Gödel’s theorem precludes consciousness being realizable on a Turing machine. Rather if I recall correctly he points out that as a possible argument and then attempts to refute it.
On the other hand Penrose is a prominent believer that human’s ability to understand Gödel’s theorem indicates consciousness can’t be realized on a Turing machine but there’s far from universal agreement on that point.
per halting problem: any system capable of self reference has unprovable (un)truths, the system can not be complete and consistent. consciousness falls under this umbrella
I'll try and ask OG q more clearly: why would the brain, consciousness, be formalizable?
I think there's a yearn view nature as adhering to an underlying model, and a contrary view that consciousness is transcendental, and I lean towards the latter
Considering how prominent gambling and gambling advertising is, aren't we creating more poverty and keeping people poor through ads? Advertising seems like it's a net drain on the poor through encouraging consumption people can't afford and pushing a variety of vices.
Edit: "Sorry your husband lost the money you were saving for a house on stake.com, but here's your free Google search."
The gains from MoE is that you can have a large model that's efficient, it lets you decouple #params and computation cost. I don't see how anthropomorphizing MoE <-> brain affords insight deeper than 'less activity means less energy used'. These are totally different systems, IMO this shallow comparison muddies the water and does a disservice to each field of study. There's been loads of research showing there's redundancy in MoE models, ie cerebras has a paper[1] where they selectively prune half the experts with minimal loss across domains -- I'm not sure you could disable half the brain and notice a stupefying difference.
> I don't see how anthropomorphizing MoE <-> brain affords insight deeper than 'less activity means less energy used'.
I'm not saying it is a perfect analogy, but it is by far the most familiar one for people to describe what sparse activation means. I'm no big fan of over-reliance on biological metaphor in this field, but I think this is skewing a bit on the pedantic side.
re: your second comment about pruning, not to get in the weeds but I think there have been a few unique cases where people did lose some of their brain and the brain essentially routed around it.
You're saying it like hardware and software are disjoint. You design hardware with software in mind (and vice versa); you need to if you want performance rivaling nvidia. This codesign, seeing their products are not only usable but actually tailored to maximize resource utilization in real workloads (not driven by w/e benchmarks), is where AMD seems to lack.
Why oversimplify the premise and frame your take as some 'proof'. Just use the term counter-argument/example
> stuff that's been removed as a "feature" isn't always stuff that nobody wants.
Graphene isn't made to cater to what everyone wants. Face ID and fingerprint unlocking so clearly have no place in a hardened OS. "Google OS-level integration is absent" should not be suprising.
This said, you ought to be able to have BFU security with stock Android and it's embarrassing Google ships stock vulnerable.
> Graphene isn't made to cater to what everyone wants.
I know! My entire point is Graphene wouldn't be a good choice for the stock OS on a mass-market phone. The Graphene devices will be great, but if Google were to replace their stock OS with Graphene there would be problems.
Virtually every issue I have with GrapheneOS stems directly from the lack of Google Play Integrity causing app incompatibilities. There's some little bits of friction here and there like security mitigations causing app crashes, but when that happens the OS tells you exactly what happened, why, and how to prevent it in the future (there's toggles to disable specific mitigations on a per-app basis). If the OS was deployed widely, those crashes would likely disappear as patches get deployed by developers.
It's very polished and completely usable as a daily driver.
If the general public prefers unsafe phones, they can chose literally any else brand. This is never going to be a mass market phone because of the tradeoffs that are perfectly fine for the intended recipients (eg people who believe a torch/calculator app REALLY doesn't need internet access, or that their Instagram REALLY doesn't need to have access to ALL the photos/videos.
Fingerprint is present in GrapheneOS. Face unlock and pattern unlock are left out because insecure. Patterns unlock is insecure in design. You start at a certain point and the next points you can go to are very limited (not the same point again and you have to be able to reach it). This makes it hard to make a strong lock. Face unlock is insecure because lack of proper hardware for it on the supported phones. Fingerprint is secure. Coercion can be worked around via 2FA feature (fingerprint + pass/PIN).
Aye but it is good Apple is safe out of the box. BFU is a low bar, and the shame is on Google.
>Lots more devices are safe BFU than just Apple's
Really? Secure against the exploits and methods these tools 3 letter agencies employ? I hate to cry source, but base Android isn't secure. What devices have similar hardware-level security, or have their Android flavor shipping with these Graphene-OS-level patches?
> Really? Secure against the exploits and methods these tools 3 letter agencies employ?
Before First Unlock data on your device is as safe as your password safe. It doesn't really matter if you use Android, iOS or any other devices as long as it have modern crypto on it.
Only being able to run transformers is a silly concept, because attention consists of two matrix multiplications, which are the standard operation in feed forward and convolutional layers. Basically, you get transformers for free.
It's kind of a shortcut answer by now. Especially for anything that touches pretraining.
"Why aren't we doing X?", where X is a thing that sounds sensible, seems like it would help, and does indeed help, and there's even a paper here proving that it helps.
The answer is: check the paper, it says there on page 12 in a throwaway line that they used 3 times the compute for the new method than for the controls. And the gain was +4%.
A lot of promising things are resource hogs, and there are too many better things to burn the GPU-hours on.
Blue-white LEDs have become the replacement for High Pressure Sodium [HPS] traffic lights because that's what the LED light companies have to sell. In the early years of the transition to LED streetlights they had to sell blue-white LED streetlights because warmer LEDs were not competitive with HPS on the basis of lumens-per-watt.
Most of the people who understood the advantages of blue-free amber HPS light over white metal halide lights retired, and this little tidbit of information didn't get passed to the next generation of city employees.
> and because people do not care.
People care, but they don't know why they hate the blue-white LED replacement lights. I've complained to the city about their new lights, but have not gotten any responses about why they haven't deployed LED lights with a safe spectrum of color.
Blue light is safer for cars - it gives slightly faster reaction times, and lower the chance of drivers falling asleep.
The problem is that for pedestrians, the reaction-time is irrelevant, they're butt-ugly, and plenty of people go on night walks because they can't sleep but want to.
Half of this article is basically about cities being overly car-centric.
In some ways, car-centric cities are like that because people don't care. "I'll just take my car, whatever". They don't care about traffic, pollution, accidents, etc.
Or maybe they care about getting to their destination in time and not wasting half a day just going from point A to point B on foot/by (shitty) public transportation, or they'll be fired.
Daily reminder that we live under capitalism where you're not allowed to just "take your time".
I’m not well-read on the old lighting research, but I’ve come across some explanations for why humans actually do much better with amber outdoor lighting than white. One of these points relates to how our pupils expand and contract with the amount of light available.
My impression was that HPS lighting became so widespread not because of the supposed advantages of its light spectrum, but because it was simply the most light-efficient technology available at that time. Here in Germany, only main/multilane streets requiring more lighting were using HPS, residential streets mostly had lamps with white fluorescent lights, so switching those to LEDs wasn't as much of a change. But still, I'm wondering: what about curtains, window blinds etc.? It's not as if people are forced to endure the intrusion of street lighting into their bedrooms?
> But still, I'm wondering: what about curtains, window blinds etc.? It's not as if people are forced to endure the intrusion of street lighting into their bedrooms?
Of course. But that's the problem -- now black-out curtains are required. And maybe you hate those because you really enjoy waking up with the sun streaming in, and now you have to wake up every morning in blackness until you go open the curtains.
The onus shouldn't be on the residents. It's the same as saying, sure it's noisy but why don't you just wear noise-canceling headphones all day long?
Government services exist to serve the people, not make the people work around them.
Low pressure sodium lights were more efficient, but they emit a single wavelength of orange light. These lights were strongly disliked by people who liked to admire their car in the streetlights (I suppose).
> residential streets mostly had lamps with white fluorescent lights,
… they used CFLs? The spiral fluorescents were invented in the 1980’s, I guess. I speculate the residential street lights used mercury vapor bulbs, which had a longer expected lifespan than fluorescents.
> But still, I'm wondering: what about curtains, window blinds etc.?
You need a good blackout curtain to deal with light pollution through your window.
Decades ago, my junior college's parking lot was lit by low-pressure sodium lighting. I recall the light being an absolute monochrome yellow, to the point that you had to be careful to remember exactly where you parked your car, because you weren't going to find it by color.
I can't vouch for Germany, but there used to be long, high-output fluorescent tubes and fixtures for street lighting in the US. They seem to have largely disappeared by the 1980s. They weren't very common, but some cities used them. They tended to be used on main streets when I saw them.
