Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Neurons unexpectedly encode information in the timing of their firing

quantamagazine.org

Post reply on HN
Print view View original

81–90 of 116 posts

Re: Neurons unexpectedly encode information in the timing of their firing

#81

Post by trhway »

eloff wrote: post #70
sdenton4 wrote: post #61

Earlier quoted context omitted.

Hilarious. We shouldn't trust rat/mouse studies, but it's typically because the mice are way LESS complicated than humans. I would expect complicated behavior in mice to be treated as a lower bound for the complexity in humans. (but what do i know, i'm just a monkey.)

There are lots of animals that can beat us at specific neural tasks. So I don't think it's helpful to think in those terms. For example, our visual short term memory is bested by chimpanzees, at least on certain tasks used to measure it across both chimps and humans. Lowly mice likely have better olfactory capabilities than us. It wouldn't surprise me if their brains can handle some very specific things better than w…

>There are lots of animals that can beat us at specific neural tasks.

anecdotally - in front of us a woman with a dog is leaving the dog park, and the dog pulls in the opposite direction that the woman tries to go, and that goes for a few seconds until the woman "Oh!, you're right, today we parked there" and follows the dog.

And on 2 occasions spread in time and space i saw a racoon confidently crossing an intensive traffic boulevard (once it was Page Mill in Palo Alto and another was Geary in SF. The Geary was around 6-7pm, getting dark, high traffic, such crossing presents some cognitive challenge even for humans) following the basic procedure we are taught in childhood - check for traffic (and the racoons were checking for the correct direction), wait for the sufficient opening, cross to the middle, check and wait again, cross.

Re: Neurons unexpectedly encode information in the timing of their firing

#82

Post by smegger001 »

smegger001 wrote: post #76
mikewarot wrote: post #64

Earlier quoted context omitted.

Quantum computers are not Turing machines as far a I know. They can encode exponentially many states simultaneously, and are non-deterministic.

they are they just dont work in binary bits they work with qbits which simply have behavior rules diffrent from digital but can be simulated by one just inefficiently. much like a binary computer can be approximated by a balanced trianay computer or analog computer or by pumps valves and flowing water like in MONIAC. they are all turring complete and can simulate each other just not efficiently. they dont have any ma…

not sure why I am being down voted here perhaps someone needs to do some reading

https://en.wikipedia.org/wiki/Quantum_computing#Computabilit...

https://www.smbc-comics.com/comic/the-talk-3

Re: Neurons unexpectedly encode information in the timing of their firing

#84

Post by Robotbeat »

IIAOPSW wrote: post #80
tsimionescu wrote: post #63

Earlier quoted context omitted.

That's still extremely unlikely given all we know about quantum computers so far, and about biology. It's also important to note that quantum computers are just faster classical computers, they are still Turing machines. Many people who are looking for some non-computable element of consciousness in quantum effects in microtubules seem to forget that.

quantum computers are not just faster classical computers. Quantum algorithms have no classical counterpart. In fact the literal clock speed per quantum computation may be slower but the computation will still get done faster overall.

But there’s no computation a quantum computer can do that a classical computer can’t. In that sense, it’s still a Turing machine. It can do some class of computation faster (in principle).

Re: Neurons unexpectedly encode information in the timing of their firing

#85

Post by LeifCarrotson »

nielsole wrote: post #69
LeifCarrotson wrote: post #67

In my very different field (robotics and industrial automation), temporal coding is one of the most powerful ways to expand your IO. Nearly all PLCs, sensors, and robots make heavy use of digital IO. But this parallel interface is limited, especially with hard-wired signals, and even if you're using serial network protocols the typical fieldbus abstraction represents the network as fixed-size buffers of digital IO th…

Not sure I understood you. Did you describe PWM?

No, typical PWM is a repetitive hardware function of an embedded system, and operates at kHz or MHz. Industrial controls usually have a scan time or requested packet interval on the order of 10 ms, much to slow for PWM without dedicated peripherals. And even if you can transmit it, you still need to read it on the other side, again, high-speed IO is a dedicated thing, you won't buy step and direction outputs or encoder inputs for unused spares.

A PLC and a robot might interchange digital signals such as "In cycle", "Part present", "Faulted", "Clear of fixture", "Screw present", and "Cycle start". Hopefully the original designers also pulled a couple spare wires in case you need to add another sensor. But the old equipment is being asked to do something new, requirements now say you need to transmit, perhaps, which of dozens of part numbers to select from, or the touch point at which a sensor fired between 10 and 100mm. That's a binary signal with 6 or 7 bits, which means a $200 8-channel output card, multiconductor cable, and a $200 input card. When possible, you'd buy the fieldbus option cards from the factory and pull network cables. But in a pinch, with the more typical two spare inputs and two spare outputs, you can move some data to the time domain and work out "If aux1 is pulsed high for 680ms, the new sensor tripped at 68mm". One wire, one bit of state (0V or 24V), many values transmitted.

You could extend this concept with a clock signal, data signal, and transmit arbitrary serial data, but that's going a bit too far for most maintenance techs. A pair of timers is comprehensible.

Re: Neurons unexpectedly encode information in the timing of their firing

#86

Post by throwaway3699 »

Robotbeat wrote: post #84
IIAOPSW wrote: post #80

Earlier quoted context omitted.

quantum computers are not just faster classical computers. Quantum algorithms have no classical counterpart. In fact the literal clock speed per quantum computation may be slower but the computation will still get done faster overall.

But there’s no computation a quantum computer can do that a classical computer can’t. In that sense, it’s still a Turing machine. It can do some class of computation faster (in principle).

If we want to reach AGI within the lifetime of the universe, and quantum effects are required for consciousness [within the lifetime etc...], it stands to reason we'd need quantum computers.

Re: Neurons unexpectedly encode information in the timing of their firing

#87

Post by smegger001 »

trhway wrote: post #81
eloff wrote: post #70

Earlier quoted context omitted.

There are lots of animals that can beat us at specific neural tasks. So I don't think it's helpful to think in those terms. For example, our visual short term memory is bested by chimpanzees, at least on certain tasks used to measure it across both chimps and humans. Lowly mice likely have better olfactory capabilities than us. It wouldn't surprise me if their brains can handle some very specific things better than w…

>There are lots of animals that can beat us at specific neural tasks. anecdotally - in front of us a woman with a dog is leaving the dog park, and the dog pulls in the opposite direction that the woman tries to go, and that goes for a few seconds until the woman "Oh!, you're right, today we parked there" and follows the dog. And on 2 occasions spread in time and space i saw a racoon confidently crossing an intensive…

as a kid i had a cat that would look both ways before crossing the street.

Re: Neurons unexpectedly encode information in the timing of their firing

#88

Post by kortex »

mr_toad wrote: post #79
radicaldreamer wrote: post #6

Earlier quoted context omitted.

There’s a strong bias toward things that are model-able in neuroscience. The role of microtubules, for example, are mostly ignored even though they may explain the complexity of cognition displayed by relatively “simple” brains.

Microtubules are found in all eukaryotic cells. Even single celled organisms. If microtubules are responsible for cognition, what are brains for, and why aren’t amoeba intelligent?

Have you ever watched an amoeba do its thing? I wouldn't call it cognition, but its behavior is remarkable for something with no neurons. In fact microtubules are the exact structure posited to be responsible for where and how amoebas move their pseudopods.

https://pubmed.ncbi.nlm.nih.gov/7983169/

Re: Neurons unexpectedly encode information in the timing of their firing

#89

Post by hau »

Robotbeat wrote: post #84
IIAOPSW wrote: post #80

Earlier quoted context omitted.

quantum computers are not just faster classical computers. Quantum algorithms have no classical counterpart. In fact the literal clock speed per quantum computation may be slower but the computation will still get done faster overall.

But there’s no computation a quantum computer can do that a classical computer can’t. In that sense, it’s still a Turing machine. It can do some class of computation faster (in principle).

Richard Feynman describes such computation, that produces incorrect results on classic computer and correct one on quantum in "Simulating Physics with Computers"[0]. He follows one physical experiment detailing every step from start to finish and how it converges to incorrect result on classical computer.

[0] - https://doi.org/10.1007/BF02650179

Post reply on HN