calepayson

> popular deep artificial neural networks (lstms, llms, etc.) are highly recurrent, in which they are simulating not deep networks, but shallow networks that process information in loops many times.

Thanks for the info. Is there anything you would recommend to dive deeper into this? Books/papers/courses/etc.

> recommend not to oversimplify structure here. what you describing is only high-level structure of single part of brain (neocortex).

Nice suggestion. I added a bit to make it clear that I'm talking about the neocortex.

> 1 & 2

Totally. I don't think AI is a simple as building a Darwin Machine, much like it's not as simple as building a neural net. But I think the concept of a Darwin Machine is an interesting, and possibly important, component.

My goal with this post was to introduce folks who hadn't heard of this concept and, hopefully, get in contact with folks who had. I left out the other so I could try to focus on what matters.

> temporal dimension is important. your article is very ML-like focusing on information processing devoid of temporal dimension. if you want to draw parallels to real neurons in brain, need to explain how it fits into temporal dynamics (oscillations in neurons and circuits).

Correct me if I misunderstand, but I believe I did. The spatio-temporal firing patterns of minicolumns contain the temporal dimension. I touched on the song analogy but we can go deeper here.

Let's imagine the firing pattern of a minicolumn as a melody that fits within the period of some internal clock (I doubt there's actually a clock but I think it's a useful analogy). Each minicolumn starts "singing" its melody over and over, in time with the clock. Each clock cycle, every minicolumn is influenced by its neighbors within the network and they begin to sync up. Eventually they're all harmonizing to the same melody.

A network might propagate a bunch of different melodies at once. When they meet, the melodies "compete". Each tries to propagate to a new minicolumn and fitness is judged by other inputs to that minicolumn (think sensory) and the tendencies of that minicolumn (think memory).

I think the evolution is an incredible algorithm is because it relies as much as it does on time.

> is this competition in realm of abeyant (what you can think in principle) or current (what you think now) representations? what's the timescales and neurological basis for this?

I'm not familiar with these ideas but let me give it a shot. Feel free to jump in with more questions to help clarify.

Neural Darwinism points to structures - minicolumns, cortical columns, and interesting features of their connections - and describes one possibility for how those structures might lead to thought. In your words, I think the structures are the realm of abeyant representations while the theory describes current representations.

The neurological basis for this, the description of the abeyant representation (hope I'm getting that right), is Calvin's observations of the structure of the brain. Observations based on his and other's research.

To a large extent, neuroscience doesn't have a great through-line-story of how the brain works. For example the idea of regions of the brain responsible for specific functions - like the hippocampus for memory - doesn't exactly play nice with Karl Lashley's experimental work on memory.

What I liked most about this book is how Calvin tried to relate his theory to both structure and experimental results.

> overall, my take it is a bit ML-like talk. if it describes real neurological networks it got to be closer and stronger neurological footing.

If, by ML-like talk, you mean a bit woo-woo and hand wavy. Ya, I agree. Ideally I'd be a better writer. But I'm not, so I highly recommend the book.

It's written by an incredible neuroscientist and, so far, none of the neuroscience researchers I've given it to have expressed anything other than excitement about it. And I explicitly told them to keep an eye out for places they might disagree. One of them is currently reading it a second time right now with the goal verifying everything. If it all checks out, he plans on presenting the ideas to his lab. I'll update the post if he, or anyone in his lab, finds something that doesn't check out.

> here is some good material, if you want to dive into neuroscience. "Principles of Neurobiology", Liqun Luo, 2020 and "Fundamental Neuroscience", McGraw Hill.

Why these two textbooks? I got my B.S. in neuroscience so I feel good about the foundations. Happy to check these out if you believe they add something that many other textbooks are missing.

Great summary. Thanks for the links. These are awesome.

Thanks for pointing this out. I’ll change it when I’m at my computer.

Awesome post, thanks. I ordered society of mind.

Reminds me of when I took "The Philosophy of Cognitive Science" in college. The entire class was on AI. When I asked the professor why, she explained: "You don't understand something unless you can build it".

It's cool to learn that quote might have been because she's a fan of Minsky.

> In terms of the root idea of the mind being a huge number of concurrent agents, I think he was close to the bullseye and it very much aligns with what you wrote.

I think you're right here and I'd like to add a bit. One common mistake people make when thinking of evolution, is where in the hierarchy it takes place. In other words, they misidentify the agents by an order of magnitude.

For example, in biology I commonly see it taught that the individual is the subject of natural selection (or worse, the population).

Really, it's the gene. The beauty of evolution is that it can take an agent as simple as the gene and shape it into the litany of complex forms and functions we see all around us.

If evolution is at play in the brain, I suspect that Minsky's agents are the individual firing patterns. Like genes, the base of the hierarchy, the fundamental unit. Also like genes, they slowly build increasingly complex behaviors from the ground up. Starting before birth and continuing for most of our lives.

Man, this has me grinning like an idiot. Thanks.

> Fantastic speculation here, explains a lot, and has testable hypotheses.

Calvin is the man.

> For example, there should be a relationship between rate of learning and the physical subcolumns - we should be able to identify when a single column starts up / is fully trained / is overused

This sounds super interesting. Could you break down what you're thinking here?

> Marvin Minsky was spot on about the general idea 50 years ago, seeing the brain as a collection of 1000s of atomic operators (society of mind?)

I'm very much an amateur in this field and was under the impression that Minsky was trying to break it up, but was trying to specify each of those operations. What I find so enticing about Neural Darwinism is the lack of specification needed. Ideally, once you get the underlying process right, there's a cascade of emergent properties.

Using the example of a murmuration of starlings I picture Minsky trying to describe phase transitions between every possible murmuration state. On the other hand I see Neural Darwinism as an attempt to describe the behavior of a single starling which can then be scaled to thousands.

Let me know if that's super wrong. I've only read second hand descriptions of Minsky's ideas, so feel free to send some homework my way.

I ordered the book and I'm checking out the website rn. Looks awesome. Thanks a ton for sharing!

Honestly, no idea.

I could imagine a Darwin Machine being a "pure brain", as you put it. While we have emotions because our brains built a pure brain atop an existing and messy infrastructure.

Or, emotions could just be the subjective experience of thoughts competing.

Calvin goes deeper into things like this in the book, but I suspect emotions help intelligence to some extent insofar as they provide environmental change. It's good for the long term health of an ecosystem to shake things up so that nothing gets too stagnant.

I'm obsessed with the idea of Darwin Machines (and I think you should be too).

I've been tinkering with the idea in python but I just don't have enough ML experience.

If you, or anyone you know, is interested in Darwin Machines please reach out!

Harari’s book is awesome but I think he gets stuck too far up the hierarchy.

Trying to describe culture with stories feels like trying to describe biology with proteins. It gets us most of the way there but also adds a whole ton of complexity because it misses the fundamental nature of the system.

Humans are great at general and accurate imitation. This probably seeded a runaway evolutionary process, the result of which was tools, fire, and language.

Fire is rad but I don’t see how it explains modern civilization.

If most animals are operating on a calorie deficit and fixing this can lead to larger brains then why haven’t domesticated animals evolved towards our level of intelligence?

There’s a professor at St. Andrew’s, Kevin Lalland I think, who has done a lot of work on imitation. His theory is that that imitation is generally useful but without high enough copying fidelity culture won’t develop.

For example monkeys have been known to exhibit fashion trends (hanging a bit of grass out of their ear). But these trends fizzle out over time. Before more behaviors can be layered on top.

I think you’re right about those factors creating a good environment for this to happen and I bet they’re sufficient but not necessary. Too bad we don’t have another example :)

It all starts with imitation of some sufficient fidelity and generality that seeds a runaway evolutionary process. Language evolved to improve the fidelity of imitation much like how genes evolved pathways to minimize random mutations during copying.

Once imitation gets good enough (general and accurate) were capable of spreading behaviors (phenotypes) without having to wait for folks to be born and grow up.

Humans are unique because we can imitate behavior well enough that a Darwinian process began.

Just like the gene, human behavior seeks out niches in the environment, eventually becoming the environment. Unlike the gene behavior evolves much faster, can be inherited horizontally and is far more prone to mutation.

Unfortunately we don’t have the equivalent of Mendelian genetics for behavior. We’re left with a unit but no understanding of the forces acting on it.

Before we knew these things in biology the best “research” we could do was classification. An interesting but unpredictive exercise. The social sciences are stuck here right now.

Until we have a the equivalent we’ll be stuck with articles like this. Articles that make a herculean (and valiant) effort to describe our society without a foundation in the physical sciences. Like trying to describe the behavior of a murmuration without knowing about starlings.

I’m studying neuroscience but very interested in how ai works. I’ve read up on the old school but phrases like memory graph and energy minimization are new to me. What modern papers/articles would you recommend for folks who want to learn more?

> Go to space

Laika does not approve of this message :)

I believe the one truly unique thing humans can do is imitate complex actions with high fidelity. When we evolved this ability it lead to an explosion of behavioral complexity simply because imitated behavior satisfies all the criteria for evolution (s/o Dawkins).

We’re not unique in our ability to imitate, but we are unique (for now) in our ability to imitate well enough that imitated behaviors are stable enough to evolve.

I think we put too much weight on intelligence as the factor differentiating innovators from the rest. If we were to clone some great scientists from the past I believe would be underwhelmed.

Not because these aren’t brilliant people but because brilliance isn’t the most important factor in innovation. Rather, innovation is driven by people who are exposed to a novel environment and are just smart enough to not ignore it.

Bill Gates’ highschool was the only one in the US with a computer. Albert Einstein was working in a patent office where he saw a conga-line of patents discussing how to standardize the measurement of time across vast geographies.

Both brilliant people. Both exposed to a novel societal idea. Thankfully they were smart enough to run with it.

I'm not sure if this is one of the consensus primary causes but the most compelling explanation I've heard is capillary damage.

Red blood cells and capillaries have roughly the same diameter leading to constant friction. Your body repairs the capillaries but, given enough time, it makes mistakes. As these mistakes pile up your circulatory system, your body's supply chain, slowly gets less efficient. This leads to every cell in your body gets the energy it needs a little later, and has to hold onto waste for a little longer.