The Algebra of Neural Networks May Explain Feature Association

The Algebra of Neural Networks May Explain Feature Association

Reading Time: 1 minute

The Algebra of Neural Networks May Explain Feature Association

This one’s not for everyone, as it’s pretty dense. And ignore the title, which was clearly written by someone other than the author of the article, as this research has nothing to do with being able to perceive higher dimensions.

What it is about is the way the local and global structure of networks come together in ways that help them connect features:

We conjecture that a stimulus may be processed by binding neurons into cliques of increasingly higher dimension, as a specific class of cell assemblies, possibly to represent features of the stimulus, and by binding these cliques into cavities of increasing complexity, possibly to represent the associations between the features.

This quote is not from the article, but rather from the original research ( Essentially, what it is saying is that they may have just found a link between the way that networks identify and associate ‘features’ in the objects that they perceive. In this work, they’re talking about biological neural networks, but I think it’s roughly analogous to the kinds layers that we see in deep learning networks in machine learning. I’m no expert in that, however, so if there are experts out there willing to weigh in, that’d be great to get your thoughts.

Special thanks to Ron Serina for flagging this one for me.

Originally shared by Tom Eigelsbach

“The progression of activity through the brain resembles a multi-dimensional sandcastle that materializes out of the sand and then disintegrates.”


  1. To QUOTE ….The PROGRESSION of ACTIVITY through the BRAIN resembles…” A MULTI-DIMENSIONAL SANDCASTLE that MATERIALISES OUT of SAND and then DISINTEGRATES…couldn’t Help myself laughing Hysterically so musingly true…oh so so Love it…!

  2. Mars channels, why not? The algebra of neural networks. The Monster dreams it all up.

  3. Gideon Rosenblatt but what neural complexity standard will a digital brain model mimick and meet? Since we can agree with the fact — the more complex an individual’s system, the more advanced operating standard that person manifests. The viability level and survival potential a homo sapience is capable of, directly reflects how neurally developed this particular brain is.

    It is a clear truth that people around have different mental agility and functionality status, especially, when they hit their ‘retirement age’. Even the number of languages a person has learned during his/her life-time shows that considerable difference in task performing efficiency.

    So, how is that Golden Standard defined for the brain model construction? Will it be more on philosophy side or on sciences side? or in Arts, or in Militarily applied knowledge? Or multilingual?

    Thank you for your another exciting subject and a branch of AI to focus on!

  4. Higher order wave forms may indeed synchronize uniform discontinuities by regulating the quantum timing of neural processing. Some Brain scientists have even suggested that electron spin may make a brain the 1st quantum computer. I still find it odd that people want to compare human sentience to a mere binary process.

  5. Phillip Johnson Your prose isn’t uninteresting, there’s a strangeness to how in the midst of many sentences that buy the benefit of the doubt with more or less brilliance, there come unique points where an implied invitation to concur to your good worded judgment meets a demise just as it seizes on terrain with which I am more familiar.

  6. Quantum computing, and cryptography in human consciousness is very new. It’s the future, just need to adapt current systems to work with the new programming language and applications. It’s like a whole new operating system. We’ll start with 0’s 1’s(current standard) and add 2’s and 3’s. Old software, just new converting it to the hardware here. Beta testers can apply.

  7. What I get out of this is that Jimi Hendrix did not die, but just went back to where he was before and after here.

  8. Boris Borcic As my notifications are now subject to erasure on a time expiration basis, if I want to further explore the errors of my literary metrics I need to do so now. I’m interested in how badly I framed the argument, (NOT MINE PER SAY) that the brain is not just a binary machine. Please respond only if you are so inclined. I am interested in where my logic about the chemical brain existing as a separate non binary quantum state failed.

  9. Phillip Johnson Sorry, I re-read your comment and can’t figure out what turn of sentence prompted my reflection more in particular. Perhaps the apparent painting of quantum computation as a “mere binary process”. It’s not: qubits are to binary like the unit circle of complex numbers is to the pair of unit real numbers, -1 and 1. But there are other junctures where your semantics isn’t rock solid. I suspect this comment of yours was notified to me with a string of others and that the others participated to the build up of my perception.

    Mmm… on second pass, I see that if this was the cause, it was grounded in a misunderstanding …and the cause of the misunderstanding was then in turn your mentioning the idea of a “mere binary process” in a context where nobody had suggested it. Also, the whole internet and all what we know computers for, including NN, are “mere binary processes” in this sense.

  10. Boris Borcic I’m going to miss this format for discussion as I’m not planning to carry with the same degree of my G+ addiction over into another social media.

    I had suspected that my badly framed statement about electron spin reversals was not a significant part of a quantum state vector in neural processing. Truth be known (And often unknowable) ambiguities are often just a matter of tone.

    I myself being wired for sound I often sound in print to the eye’s ears, like the cowardly braggart that I am. Fawning for attention, all of my life I’ve been attacked for telling the truth. I often speak (and write) with the frail conviction of a weak authority or even a fraud. Were I not a showoff and a bad one at that, I might never have learned the cultivated appetite for engaging the critique. Again your critiques inspired me to write the short fiction piece of “When Spam Bots Fall In Love.”

  11. Phillip Johnson what serendipitous stereopsis can spambots of distinct inspirations or training — spambots with parallax — achieve when mutually in love?

  12. I guess we could suspect a mutual dyad of singularity. It’s only Spi Fi.(Together we learn as one?)

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Sign up here for the latest articles. You can opt out at any time.

Subscribe by email:

Or subscribe by RSS:

%d bloggers like this: