I’m interested in getting people reactions to this paper: The world as a neural network.
Here’s a fragment of the abstract that omits the technical parts:
We discuss a possibility that the entire universe on its most fundamental level is a neural network…This shows that the learning dynamics of a neural network can indeed exhibit approximate behaviors described by both quantum mechanics and general relativity. We also discuss a possibility that the two descriptions are holographic duals of each other.
Here’s an article and interview with the author.
I don’t understand the math and I’m skeptical of another Theory of Everything, but there are few items that struck a cord with me.
For one thing, it tries to tie learning to entropy and to view the evolution of the universe as terms of learning. This includes, of course, the physical aspects of the universe, but also biological evolution, as a form of learning. Consciousness and its association with learning emerges directly from the overall processes.
Indeed, if the entire universe is a neural network, then something like natural selection might be happening on all scales from cosmological (> 10+15 m) and biological (10+2 − 10−6 m) all the way to subatomic (< 10−15 m) scales. The main idea is that some local structures (or architectures) of neural networks are more stable against external perturbations (i.e. interactions with the rest of the network) than other local structures. As a result the more stable structures are more likely to survive and the less stable structures are more likely to be exterminated. There is no reason to expect that this process might stop at a fixed time or might be confined to a fixed scale and so the evolution must continue indefinitely and on all scales. We have already seen that on the smallest scales the learning evolution is likely to produce structures of a very low complexity (i.e. second law of learning) such as one dimensional chains of neurons, but this might just be the beginning. As the learning progresses these chains can chop off loops, form junctions and according to natural selection the more stable structures would survive. If correct, then what we now call atoms and particles might actually be the outcomes of a long evolution starting from some very low complexity structures and what we now call macroscopic observers and biological cells might be the outcome of an even longer evolution. Of course, at present the claim that natural selection may be relevant on all scales is very speculative, but it seems that neural networks do offer an interesting new perspective on the problem of observers.
Broad Speculations 
One thing occurred to me after publishing this was when we look at the universe we might, in a sense, just looking at ourselves in the mirror. If our brain can be described as a neural network, are we just seeing our own reflection if the universe can be described the same way?
LikeLiked by 1 person
I haven’t read the article. I suspect it is wishful thinking but I wonder about the language being used. For example, “We discuss a possibility that the entire universe on its most fundamental level is a neural network” … uh, like a neural network? A neural network requires nerves, no?
I suggest this is follow as the universe is vast and the ability to communicate between parts only at the speed of light (vacuum) or less is very limiting.
There seems to be an innate desire on the part of some for the universe to be a whole thing, rather than a collection of parts. People want the whole universe to possess consciousness, for example. This may be rooted in wanting to be a small part of something bigger, I do not know (but suspect).
The universe being able to “learn” makes it more like a living entity, no?
Am I shocked that one complex system may have things in common with other complex systems? No. But having things in common does not make them the same or even alike, except in small ways.
LikeLiked by 1 person
In AI there are neural networks that supposedly are intended to work like neural networks in the brain. But it turns out even those neural networks aren’t completely understood. The author writes.
The entire universe on its most fundamental level is a neural network. This is a very bold claim. We are not just saying that the artificial neural networks can be useful for analyzing physical systems] or for discovering physical laws, we are saying that this is how the world around us actually works. With this respect it could be considered as a proposal for the theory of everything, and as such it should be easy to prove it wrong. All that is needed is to find a physical phenomenon which cannot be described by neural networks. Unfortunately (or fortunately) it is easier said than done. It turns out that the dynamics of neural networks is so complex that one can only understand it in very specific limits.
LikeLike
I’m also skeptical of ‘Theory of Everything’ solutions, not least because the mathematician in me accepts ‘undecidability’ and ‘incompleteness’ as facts of life. (We may never know.) But there’s a certain pleasant – almost ‘scientific Buddhism’ – about the idea. Some overlap with my sci-fi novel too of course.
LikeLiked by 1 person
That sort of pleasant – almost ‘scientific Buddhism’ was what resonated with me too.
LikeLiked by 1 person
Um… no. (Sounds like science fiction to me.)
LikeLiked by 1 person
I think, if there is anything to the proposal, it is the idea there may be some sort of basic laws governing entropy, evolution, and learning that might apply to non-organic and organic world. These laws in operation may resemble neural networks and be usefully described in that way.
LikeLiked by 1 person
No harm in speculating! 🙂
LikeLiked by 1 person
Isn’t that what we do all the time?
LikeLiked by 1 person
This blog is called “Broad Speculations” for a reason. 🙂
LikeLike
BTW, there is a quantum Darwinism theory.
https://en.wikipedia.org/wiki/Quantum_Darwinism
LikeLike
Does he address the speed of light limit anywhere? If not, then if the universe is a neural network, it’s a newborn brain dissolving before it has a chance to have a coherent thought. Due to the ongoing expansion of the universe, a thought that begins right here could only ever affect 3% of the observable universe.
I think the idea of quantum Darwinism is interesting, but I’ve always struggled to find a clear non-superficial description of how it’s supposed to work.
LikeLiked by 1 person
There are some light-cone statements in relation to some equations but that’s in math I’m not going to attempt to understand. Since he thinks relativity can be derived from his approach, it some way it must address the speed of light I would think.
My interpretation is that there is no “thoughts” involved in any of this. I don’t think he is claiming the universe is some sort of enormous brain.
If you think of the diversity and complexity of life on Earth evolving from single cell organisms, then think of all of matter, space-time, gravity, and everything else evolving in an analogous way from more fundamental constituents in a way governed by processes similar to neural networks.
LikeLiked by 2 people
Thanks for the clarification. If the assertion is that computation happens everywhere, then I’m on board with that. But saying it’s all a neural network seems pretty specific, implying everything has a functional topology similar to a neuron. I don’t know. It seems like an idea that would require a lot of interpretational energy. If his paper is that mathematical, I think I’ll let the physics community assess whether this is anything worth investing time in.
LikeLike
I think the physics community will not think highly of it. But I thought it interesting because it aligns to a good degree with some of my more speculative speculations in some of these posts.
More generally the idea is the universe wasn’t essentially created at the Big Bang but may have been evolving possibly both before and after the Big Bang in some way driven by entropic processes. The evolution of life and consciousness is a part of larger evolving process.
LikeLiked by 2 people