Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Scientists Discover Exotic New Patterns of Synchronization (quantamagazine.org)
83 points by headalgorithm on April 8, 2019 | hide | past | favorite | 20 comments


This bit:

"The scientists mapped the different synchronized clusters that can form in a network of oscillators to that network’s symmetries. In this context, symmetries refer to the ways a network’s oscillators can be swapped without changing the network, just as a square can be rotated 90 degrees or reflected horizontally, vertically or diagonally without changing its appearance." ... "The synchronized part is one big synchronized cluster, and the desynchronized part is a bunch of single clusters"

Sounds like Chladni/Faraday patterns to me:

https://en.wikipedia.org/wiki/Ernst_Chladni#Chladni_figures

The symmetry group might be the network of anti-nodes, which vibrate in common with the shared energy reflecting through the medium. The nodes seem less likely, as that's where the energy waves cancel out, thus providing less kick against the local oscillators to overcome their independent motion.


I wonder what those would look like if you had multiple oscillators that could move around the plate along the gradient of lowest resistance to their oscillation, sort of like the opposite of the flour. I'm also curious if there's a way of extending (or reducing?) this bulk material resonance to graphs of incompletely interconnected nodes.


In a Mythbusters episode they set up coupled metronomes. One of the metronomes didnt sync up and the host was annoyed. He may have unwittingly discovered the chimera state https://www.youtube.com/watch?v=e-c6S6SdkPo


>Roukes, who is a professor of physics, applied physics and biological engineering at Caltech, is most interested in what the ring of NEMs suggests about huge networks like the brain. “This is very, very primordial compared to the complexity of the brain,” he said. “If we already see this explosion in complexity [in a network of 8 NEM oscillators], then it seems feasible to me that a network of 200 billion nodes and 2,000 trillion [connections] would have enough complexity to sustain consciousness.”

I wonder if there's a good way to integrate something about oscillation into neural networks.


You may be interested in György Buzsáki’s “Rhythms of the Brain”:

https://www.amazon.com/Rhythms-Brain-Gyorgy-Buzsaki/dp/01998...

“...self-emerged oscillatory timing is the brain's fundamental organizer of neuronal information. The small-world-like connectivity of the cerebral cortex allows for global computation on multiple spatial and temporal scales. The perpetual interactions among the multiple network oscillators keep cortical systems in a highly sensitive "metastable" state and provide energy-efficient synchronizing mechanisms via weak links.”


"Connectome-specific harmonic waves" maybe: https://www.nature.com/articles/ncomms10340


Here is also a Nature paper about using synchronization phenomena in coupled solid state spintronic nano-oscillators to achieve vowel recognition:

https://www.nature.com/articles/s41586-018-0632-y

Arxiv PDF: https://arxiv.org/ftp/arxiv/papers/1711/1711.02704.pdf


Sci-hub doesn’t seem to have it, but Google shows the full paper mentions something about simulation. It looks like it’s only 8 oscillators and they actually built a physical implementation — why couldn’t this all have been done in simulation?


I think it would be fair to say that the entire purpose here is to determine exactly what it is we would need to understand and do to create an effective simulation. Hard to simulate when you're not sure what the relevant parameters are in the first place. (I say "hard" and not "impossible", because simulations can be a relevant part of working out what those parameters are, but speaking generally, that's something that you can do when you're already 90% of the way there, and it doesn't sound like we necessarily are.)


http://scholar.google.com has the full-text pdf if you put in the title.


Ah, thanks! PDF link:

http://nano.caltech.edu/publications/papers/Matheny-2019-Sci...

Looks like they did simulate it, and the experimental results differed from the simulation due to number of harmonic terms used: “To obtain quantitative agreement between the phase model (Eq. 3) and the experiment, additional terms in the harmonics of the magnitudes would need to be included in the phase model.” Which I suppose implies that very weak, distant harmonics might play an important role in these patterns.

Also an earlier paper from the same group that discusses infeasibility of other approaches: https://authors.library.caltech.edu/81293/1/acs.nanolett.7b0...


Sci-Hub does have the Science paper: http://sci-hub.tw/10.1126/science.aav7932


There are awesome videos of ranks of metronomes syncing up.

I wonder how synchronization "interacts" with Relativistic non-simultaneity?


Relativity (even of the ’easy’ Special variety) can have very peculiar effects on physical systems.... I remember trying to solve for a relativistic pendulum, and it was a seriously intricate situation.


Several years ago at DEFCON the badges had LEDs but also had IR LEDs, as well as an IR sensor. I wanted to write a program to flash the LEDs, but wanted groups of the badges to synchronise their flashes... It was harder than I thought, and I never did come up with an appropriate solution, though I think about it from time to time.


May be you hadn't come across https://ncase.me/fireflies

If you look at the news.yc discussion https://news.ycombinator.com/item?id=14452832 at least one user had managed to sync LEDs using a similar model.


Well gee... That probably would have worked... maybe I'll try it again next time I get the chance.


DC22? iirc those ir LEDs were pretty directional


My favorite was the unwieldy vinyl badge and seeing people try to figure out how to attach it to the lanyard...

My friend almost lost his at the John MacAfee party when it fell off the lanyard.


I read dentists. ..




Consider applying for YC's Fall 2026 batch! Applications are open till July 27.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: