Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Map of an Insect’s Brain

smithsonianmag.com

Post reply on HN
Print view View original

101–110 of 128 posts

Re: Map of an Insect’s Brain

#101

Post by abecedarius »

stevenjgarner wrote: post #98

Is it reasonable to expect that as this science progresses, it will progressively map out more complex connectomes? So the connectomes of nematodes and sea squirts have only a few hundred neurons each; and a fruit fly larva has 3,016 neurons with 548,000 synapses; and humans have 86 billion neurons with hundreds of trillions of synapses; is the expectation that brain maps will progressively be made of animals that ha…

"Bees: a new unit of measurement for ML model size" https://www.lesswrong.com/posts/YKfNZAmiLdepDngwi/gpt-175bee

(lists some other animals too)

Re: Map of an Insect’s Brain

#102

Post by toss1 »

anonymouskimmer wrote: post #79
toss1 wrote: post #75

Earlier quoted context omitted.

Seems two different things are being discussed, sort of top-down, vs bottom-up modelling. I'm sure we're not far from making a high-level LLM-ish model of behavior based on those extensive studies. But the topic of discussion is not that, but making a model sufficiently accurate that you "turn the crank" and it yields similar behavior without any priors of what that behavior should be. To do that, at a minimum, we'd…

> But the topic of discussion is not that, but making a model sufficiently accurate that you "turn the crank" and it yields similar behavior without any priors of what that behavior should be. Change the expression levels of some genes, alter an amino acid here or there, change the input parameters (e.g. make a phenotype that lacks the ability to feel pain), and you can end up with a neuron network that responds subt…

Exactly! The rabbit hole of determinism just goes deeper and deeper, the closer we look. Kind of turtles all the way down...

Re: Map of an Insect’s Brain

#103

Post by toss1 »

epgui wrote: post #77
toss1 wrote: post #75

Earlier quoted context omitted.

Seems two different things are being discussed, sort of top-down, vs bottom-up modelling. I'm sure we're not far from making a high-level LLM-ish model of behavior based on those extensive studies. But the topic of discussion is not that, but making a model sufficiently accurate that you "turn the crank" and it yields similar behavior without any priors of what that behavior should be. To do that, at a minimum, we'd…

Edit: I got confused and didn’t write this comment in the right place. —- I somewhat agree with how you described it as top-down vs bottom-up. I think it’s not exactly how I was framing it, but it’s close enough, and it’s a useful way to think of it. Even in the rest of your comment you’re taking a bit more of a bottom-up approach relative to what I’m saying: you’d be surprised how much we know about how the brain’s…

Thanks for the recommendation, tracking it down.

And yes, modeling at the higher functional level can be very useful; knock out the Wernicke's center and speech goes, visual cortex, vision, etc... So, with a more detailed functional description of each level, we may wind up with a model with useful predictive value.

Tho, that said, how does this approach create a truly robust abstraction from the lower level wetware? Would it provide ability to fully reproduce computations? Would it account for lower-level changes in health, hormones, electrolyte levels, neurotransmitter-active drugs...?

Re: Map of an Insect’s Brain

#104

Post by abecedarius »

abraxas wrote: post #5

Is it simulated to the level where you could give it some visual stimulus and observe the actions it is trying to take? Could it be wired to a remote robotic insect and control it in real time? I've no idea how detailed these simulation projects and if we are months or decades away from doing what I mentioned

Good overview of what that sort of thing would take and what was known and unknown: https://www.fhi.ox.ac.uk/brain-emulation-roadmap-report.pdf

(2008. Of course there's been related work since, but I don't know of any attempt to review all the newer stuff together.)

Re: Map of an Insect’s Brain

#105

Post by DonHopkins »

pmoriarty wrote: post #60
DonHopkins wrote: post #36

Earlier quoted context omitted.

In Jaron Lanier's review of John Markoff's book "What the Dormouse Said", he mentioned an exchange between Douglass Engelbart and Marvin Minsky: https://web.archive.org/web/20110312232514/https://www.ameri... >Engelbart once told me a story that illustrates the conflict succinctly. He met Marvin Minsky — one of the founders of the field of AI — and Minsky told him how the AI lab would create intelligent machines. Eng…

Intelligent machines can't be of benefit to people? I have difficulty believing Englebart could have been so short-sighted.

I have no problem believing you totally missed the point.

Re: Map of an Insect’s Brain

#106

Post by flangola7 »

yourapostasy wrote: post #95
Groxx wrote: post #46

Earlier quoted context omitted.

For some more context / concrete links: there are ongoing efforts to simulate the C. elegans worm, e.g. https://openworm.org/ , which has ~300 neurons. The actual precision of this model is: nobody knows, because nobody knows precisely what neurons do / what they react to. We know some of it but definitely not all. But, simulating what we do know, you get quite worm-like behavior, despite whatever flaws exist. To get…

An average cell contains an estimated 100 trillion atoms. I’m guessing to “just” simulate elegans’ 302 neurons at the atomic not to speak of quantum precision, is not within our compute capabilities yet. Until then, we’ll have to content ourselves with rather crude approximations of simulations. I’m not entirely convinced on a gut feel level that the current GPT (multimodal or not) AI will necessarily lead to AGI whe…

I think it's likely we will figure out how to make intelligence in silicon before we figure out how it works in brain matter.

Just as there's more than one way of generating electricity there may be many ways of generating thought.

Re: Map of an Insect’s Brain

#107

Post by jokoon »

jll29 wrote: post #39
jokoon wrote: post #28

I wish AI scientists would try to work on this instead of machine learning.

Machine learning has much more humble goals: fitting a bunch of data to a curve. Despite all the hype around chatGPT, I have yet to see any model that asks me a question (without being programmed to do so). Today, my son asked me out of the blue: "Why do people write on paper?" and "What are our walls made of?" and "Why don't we paint our house yellow?". I don't care to live extraordinarily long, but I'd give my righ…

that anecdote about a child asking question is a really impressively relevant point about AI

Re: Map of an Insect’s Brain

#108

Post by seesawtron »

xipho wrote: post #49
causality0 wrote: post #7

With 3,016 neurons and 548,000 connections, called synapses, the result is by far the most complex map of a whole brain ever made. It's an impressive achievement but I'm not sure I'd call that a whole brain. It's a larval proto-brain with less than 4% the neurons of an adult fly.

Science and Nature are the canonical exaggerators when it comes to headlines. Whole genome, whole mitochondria, complete, total, you name it, there is (nearly?) always an invisible asterisk at the end. When these words are use then Materials and Methods tell you what actually happened. What's completely crazy in this research is the ability to thin-section fly brains. Thousands and thousands of slices _of a fly brain…

This is an electron microscopy based study, not light microscopy (i.e. SPIM or open-spim). Or did you mean something else?

Re: Map of an Insect’s Brain

#109

Post by seesawtron »

FrustratedMonky wrote: post #18
notfed wrote: post #14

Earlier quoted context omitted.

In this case the "map" is simply a set of high-res images, not a computer model. I don't think an image can be aware.

Guess I need to find source study. This high level article makes it sound like an actual map, not an image. Based on same type of map that has been made of nematode From article Now, researchers have constructed a detailed map of the neurons and the connections between them in the brain of a larval fruit fly. With 3,016 neurons and 548,000 connections, called synapses

It's just a graph of connections with neurons as nodes and synapses as the connections. Poorly translated as a "map" in reference to the historical studies that have tried to "map out" the brain wiring.

Re: Map of an Insect’s Brain

#110

Post by wizofaus »

FrustratedMonky wrote: post #89
wizofaus wrote: post #25

Earlier quoted context omitted.

Is there convincing evidence actual insects have "awareness"?

There isn't convincing evidence that humans have 'awareness'.

My awareness is the only thing I have 100% convincing evidence for the existence of!
Post reply on HN