_______ __ _______
| | |.---.-..----.| |--..-----..----. | | |.-----..--.--.--..-----.
| || _ || __|| < | -__|| _| | || -__|| | | ||__ --|
|___|___||___._||____||__|__||_____||__| |__|____||_____||________||_____|
on Gopher (inofficial)
URI Visit Hacker News on the Web
COMMENT PAGE FOR:
URI A Eureka machine that thinks like nature and explores what AI cannot
ninjagoo wrote 3 hours 42 min ago:
Seems like the key elements in this are the use of a neuromorphic
autoencoder (instead of a 'regular' one), plus FowlerâNordheim
annealing dynamics and Ising energy minimization so that the system is
not just passively settling, it's being taken through a controlled
search process designed to avoid premature trapping and scale to
higher-order combinatorial optimization problems. [1] A 'regular'
autoencoder is a neural network trained to compress data and then
reconstruct it.
A neuromorphic autoencoder is instead implemented using brain-inspired
computing elements like spiking neurons, event-driven updates, local
interactions, sometimes specialized hardware. In this paper, looks like
the autoencoder is being used as a structured energy-minimizing circuit
for an Ising optimization problem. The architecture manipulates Ising
clauses rather than only pairwise spin interactions.
Ordinary artificial neurons compute matrix ops such as y=f(Wx+b), while
this uses artificial neurons that accumulate input, which emit a spike
when they cross a threshold, like biological neurons (event driven
neural dynamics).
URI [1]: https://www.nature.com/articles/s41467-026-71937-4
K0balt wrote 6 hours 11 min ago:
I wonder how this compares to thermal wells? [1] They seem to work in a
similar way, sampling from chaotic datasets to find the lowest energy
state.
Is one fundamentally more scalable? More efficient?
URI [1]: https://extropic.ai
adrianwaj wrote 2 hours 30 min ago:
Computers that don't require cooling - so cool - literally!
stared wrote 6 hours 28 min ago:
The whole title is a buzzword cluster, until proven otherwise.
Which tasks, in particular, does it do better? Not as in "it could do
them better", but actually there are benchmarks. If they are, they are
buried beneath marketing; if not - well, we have our answer.
What is "thinks like nature"? Spin systems, are no more (or less)
nature than transistors.
That said, I am all for exploring various systems for computation and
simulation - I think there is a lot to discover.
otikik wrote 3 hours 21 min ago:
Yeah, I was reminded of the Rockwell Retro Encabulator a little bit.
URI [1]: https://www.youtube.com/watch?v=RXJKdh1KZ0w
ngriffiths wrote 4 hours 44 min ago:
Yeah, I mean it's obviously meant to be a marketing pitch but it's
not a very good one.
> The hardest computational problems are not waiting for faster chips
â they are waiting for machines that compute in a fundamentally
different way.
Surely they don't actually believe that, right? Like you say the
benefits must be limited to specific shapes of problems (not all of
"the hardest" ones), and the whole history of computing is about how
faster chips is an excellent answer to difficult computational
problems.
unmole wrote 3 hours 49 min ago:
You're describing The Hardware Lottery:
URI [1]: https://arxiv.org/abs/2009.06489
ngriffiths wrote 16 min ago:
That's interesting, thanks. I only read the abstract so far but
was immediately reminded of this recent HN submission[1] and the
whole thing that certain ideas go together, and so they are
adopted together, but the resulting bundle of ideas might be
poorly suited to certain problems.
[1]
URI [1]: https://news.ycombinator.com/item?id=48237163
gobdovan wrote 4 hours 24 min ago:
> and the whole history of computing is about how faster chips is
an excellent answer to difficult computational problems.
I don't really disagree, and I am definitely not taking their
marketing pitch seriously. Yet, you could look at the same
computation history and interpret it as an economically constrained
hill-climbing around an idea that was simple enough to work
reliably (von Neumann architecture) and that worked and scaled so
well that we were rarely forced or desperate enough to move
conceptually far away from it.
Sufficiently general digital computers can simulate other
computational models, so I think 'faster' is ultimately the end
game, but for some classes of computation, as you also noted, we
may need to go for analog hardware, (maybe) quantum devices,
optical interconnects, and so on.
Bret Victor has a talk about this, more or less: [0]
[0]
URI [1]: https://www.youtube.com/watch?v=8pTEmbeENF4
b800h wrote 8 hours 33 min ago:
There are a lot of buzzwords in there. Does it work?
GistNoesis wrote 9 hours 23 min ago:
I think this is about Ising Computers. I can't judge whether or not the
worth of this paper.
But here are some good video introduction for what Ising computers are
and how they work by
Aaron Danner : [1] Ising Computers #1: Introduction
Ising Computers #2: The Number Partitioning Problem
Ising Computers #3: The Max-Cut Problem
It's an alternative way of computing, by setting up physical system,
letting them evolve, and looking what state they evolve to.
You are setting problem by defining a system of coupled harmonic
oscillators. Statistically (Boltzmann) after a long time it should
settle in a configuration of low energy state, where the energy
function is defined by the values of the coupling constant you set up.
It has a lot of similarity with quantum computing but none of the
weirdness and you can simulate them numerically on standard computer
instead of using real hardware to study them.
URI [1]: https://www.youtube.com/watch?v=mD-0VpNSJA0&list=PLXb3r5ny8_1X...
gobdovan wrote 9 hours 31 min ago:
> a neuromorphic computer that combines quantum-tunnelling physics with
a brain-inspired architecture
This ought to be the most rhetorically compressed,
stacked-legitimacy-seeking hype phrase I've ever seen in a tech
description.
fc417fc802 wrote 9 hours 21 min ago:
Amusingly the nature paper is also an incredibly dense wall of hype
terms but actually appears to have substance. It's like a weird
alternate reality where a scam artist attempting to fleece gullible
investors took things too far and performed rigorous science.
anthk wrote 9 hours 38 min ago:
We should ask Stuart Hameroff for help then.
Othrya wrote 10 hours 11 min ago:
Yes, I actually believe that if we really want to build AI and physical
AI, we need this. I'm working on this for a while. vantar.xyz
dave1010uk wrote 10 hours 15 min ago:
> Explores what AI cannot
In other words, gradient descent isn't good at combinatorial
optimisation. I'm sure the research is better but the hype in the blog
post leaves a bad taste.
There must be a version of Rich Suttonâs Bitter Lesson that applies
to alternative computing like this, along with all the other exciting
specialised hardware we've seen come and go over the years, like expert
systems, optical computing, neuromorphic computing, etc.
Something like:
General purpose commodity silicon with rapidly evolving software
generally beats specialised hardware.
Software is just so much faster to iterate and improve than hardware.
AI is also improving it too (eg AlphaEvolve).
Specialized hardware may give a single, significant improvement that
grabs headlines but in the long term, compounding small improvements
win.
thesz wrote 43 min ago:
> gradient descent isn't good at combinatorial optimisation.
If you convolve your problem with sufficiently wide Gaussian, you can
use gradient descent. The approach is called Natural Evolution
Strategies [1] [1]
It requires O(N^4) evaluations to compute Fisher Information Matrix
for N-dimensional parameterization of the problem in original
formulation. But there are closed form solutions and more economical
representations of covariance matrix (LoRA, hehe).
URI [1]: https://en.wikipedia.org/wiki/Natural_evolution_strategy#Nat...
nyeah wrote 6 hours 28 min ago:
I'm not sure whether these FPGA codes count as specialized hardware.
bitwize wrote 8 hours 9 min ago:
> General purpose commodity silicon with rapidly evolving software
generally beats specialised hardware.
All of the Amiga people are sighing right now, as they recall how
their beautiful, elegant system synergistically designed with custom
chips was outpaced by CPU/memory brute force in the early 90s.
anthk wrote 9 hours 44 min ago:
In hardware Prolog/Kanren/expert systems? That would be possible with
libre microcode for Intel, and not this spyware corporate shithole we
are living it.
We would be able to switch microcode at boot and set one for
security, another one for C performance, others for Lisp performance
and so on.
sixtyj wrote 10 hours 5 min ago:
âneuromorphic computer that combines quantum-tunnelling physics
with a brain-inspired architecture to find solutions to hard
mathematical problemsâ
I have Bruce Sterlingâs Ascendaries: The Best of Bruce Sterlingâ
and⦠the reality is somewhere here in his storiesâ¦
Or take Charles Stross and his Accelerando book.
Do you think that teams behind such projects are avid readers and
just fulfill the sci-fi stories? :)
geremiiah wrote 10 hours 12 min ago:
I don't think they are even referring to gradient descent here. I
think they are referring to systems like AlphaEvolve where they use
LLMs to give an informed/heuristical guess to try to tackle an
otherwise insurmountable search space.
ktallett wrote 10 hours 19 min ago:
They have replicated a neuromorphic algorithm (brain like) on a FPGA,
but this implementation at this scale is doubtful to have any
improvement over a brute force effort. Quite a few companies feel this
is the way forward, although the end goal would be potentially better
using photonic chips than qubits and obviously better than an fpga.
The title is especially buzzword based with minimal meaning for the
actual paper.
viccis wrote 10 hours 23 min ago:
This reads like the paper from the Sokal affair.
mrandish wrote 9 hours 51 min ago:
It really does. The verbiage just reeks of gratuitous buzzword
grandiosity.
realo wrote 10 hours 24 min ago:
So many ... words... big words ...
Can't compute.
Help.
wmertens wrote 9 hours 53 min ago:
I had a long ELI16 session with Claude about it, and the way I
understand it is that they
- use Ising machines to describe a certain problem into clauses,
storing system state (e.g. spin of something) in variables
- then use a neural network layer where each neuron determines the
value of one clause
- then for each state item, use the neuron output to determine if
flipping that state would improve the overall system score
- and then use FN-like "noise" to determine whether to flip or no
If the energy landscape of the problem is pretty local, this is
guaranteed to find a good solution to the system, using way less
compute than brute-forcing it.
me551ah wrote 10 hours 32 min ago:
This isnât even a research paper.
Is there some code or results from experiments where we can see the
speed up?
jumploops wrote 10 hours 28 min ago:
Paper is linked on the page (doi.org link redirects to Nature), code
here[0]
[0]
URI [1]: https://github.com/aimlab-wustl/NeuroSA-HO
jumploops wrote 10 hours 33 min ago:
Higher-order neuromorphic Ising machinesâautoencoders and
Fowler-Nordheim annealers are all you need for scalability[0]
[0]
URI [1]: https://www.nature.com/articles/s41467-026-71937-4
froh wrote 6 hours 58 min ago:
ah that's the actual paper the OP is about! took me a bit. thanks
for the link.
geremiiah wrote 10 hours 24 min ago:
OK, this is just ridiculous now. Cut it with all this "all you need"
crap.
I'm only commenting on the title. I like their work.
i7l wrote 7 hours 1 min ago:
The unreasonable effectiveness of regurgitated partial titles.
repelsteeltje wrote 10 hours 36 min ago:
> [...] quantum-inspired computing built on CMOS technology [...]
So at the heart of the solution is some FPGA that does something (close
to?) quantum computing and that helps exploring exponential search
space in somewhat feasible way? Is the gist that we might have stumbled
upon a practical application of QC? And if so, what's the secret sauce
if not lots of qbits? A new algorithm? Is it just hype?
Can someone that understands quantum computing please comment?
pathOf_aFineMan wrote 7 hours 47 min ago:
the use of 'quantum' appears to be tagging onto the potential of
quantum annealers (which have repetitively [1] [2] been shown to be
classically tractable) while trying to mimic a kind of quantum
tunneling, ie the annealing schedule, without any kind of promises
about exponential speedups etc. Quantum annealers themselves have few
promised advantages for general combinatorial optimization problems
without significant changes to extant hardware paradigms [3] [1] [2]
URI [1]: https://arxiv.org/abs/2503.05693
URI [2]: https://arxiv.org/pdf/2507.22117
URI [3]: https://arxiv.org/abs/2008.09913
dv_dt wrote 8 hours 8 min ago:
I'll have to see if I can find references to an older effort on
previous learning algorithm optimizations with FPGAs in the loop - it
must be 20+ years old by now. The algorithm did indeed optimize the
toy problems that it was setup to optimize, but it exploited
non-digital, analog electronic characteristics of the individual FPGA
to do it, so the solutions were not portable to any other FPGA - even
of the same model.
Edit: There it is, Adrian Thompson evolution of tone generators,
1997.
URI [1]: https://en.wikipedia.org/wiki/Evolvable_hardware
jumploops wrote 10 hours 13 min ago:
So this isn't quantum computing (in the qubit sense), but instead a
different computer architecture (demonstrated on an FPGA) that's
based on FowlerâNordheim (FN) quantum tunneling (a real physical
effect, used in flash memory, but simulated here).
From the paper:
> The FN-dynamics may be realized either by a physical FN-tunneling
device or via a digital emulation of the FN-tunneling dynamical
systems. In this work, we employ the digital emulation to achieve the
precision required for simulated annealing in the low-temperature
regime.
With a "real" (read: analog) FN device, you potentially get large
speed ups and even larger cost/energy savings, because the physics is
essentially working for "free" -- that's the quantum part.
What's unclear is how scalable the autoencoder architecture would be
with analog FN devices today.
ktallett wrote 10 hours 23 min ago:
This is not especially related to quantum computing. Neuromorphic
computing uses an algorithm that tries to replicate how the brain
works and then in this case implements it and runs it on an FPGA.
There are quite a range of papers on this concept and multiple
companies are doing just this to show their work. It is often used as
it should theoretically avoid such a brute force approach.
swiftcoder wrote 10 hours 25 min ago:
This is not quantum computing - "quantum-inspired" could just as well
be used to describe a process like simulated annealing. The problem
they are solving here is a problem often used as a benchmark for
quantum computing, but the approach is purely classical.
wmertens wrote 10 hours 27 min ago:
No it's just analogies. It's a normal FPGA.
pipo234 wrote 10 hours 35 min ago:
> Can someone that understands quantum computing please comment?
...
Crickets
...
incognito124 wrote 6 hours 23 min ago:
Very well then, could someone that _got used to quantum computing_
please comment?
DIR <- back to front page