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

51–60 of 128 posts

Re: Map of an Insect’s Brain

#51

Post by lossolo »

xyproto wrote: post #31
yyyk wrote: post #22

Earlier quoted context omitted.

Neurons and synapses are incredibly complex. Think you're emulating over 3,000 heterogeneous cores with over half a million links while communication must be low-latency-ish. A third of the links seem to join other links and we don't even know what that does. If there's computation there we'll need even more cores.

It could be simulated slowly at first.

To simulate something you need to first know how it works and know all the relevant interactions, this is something that we currently do not understand. Neurons are not equivalent of neural networks we use in computers, they are A LOT more complex with whole groups firing the same time and chemicals regulating neurons all the time and whole topology is plastic, it works in a way we can't model or simulate currently. People hugely overestimate our knowledge about brains.

Re: Map of an Insect’s Brain

#52

Post by neurohdmi »

yyyk wrote: post #8
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

>Is it simulated to the level where you could give it some visual stimulus and observe the actions it is trying to take? No way. First, this mapping doesn't tell us how the synapses are regulated - if we could 'run' this the weights would stay fixed forever and that's not how a brain works. Second, there must be some neurons dedicated to chemical management, and they'd go haywire unless you found a way to deal with t…

The first thing you said is correct. To do a proper simulation you would need to gather functional properties of the various cell classes and their synaptic connection, which this study didn't do. (Maybe you can find that information from other lab, I'm not familiar with fly models?)

However we definitely have the computational ability to do simulations a fly network. Look at some of the modeling done by the Blue Brain Project or Allen Institute for Brain Science - they do simulations of rat and mice models with hundreds-of-thousands to millions of neurons and exponentially more synapses. 3000 neurons is not that many. If you stuck to non-compartmental point models a 3,000 neuron simulation could probably be ran on a moderately high-end laptop.

But as said before, the physical connectome is only part of the information you'd need do any worthwhile simulations.

Re: Map of an Insect’s Brain

#53

Post by kelseyfrog »

gptgpp wrote: post #16
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

Man... I just find neural connectomes so depressing. It's like looking at the copper wiring on the motherboard, or the pins of the CPU, when what you really want is the logic from the networked gates (transistors). Yet it seems we are many, many decades away from being able to extract that in any comprehensible or definitive way. I need to stop reading neuroscience articles. There's always big proclamations, Like "th…

Idk, I think it's like looking at a winning lottery ticket[1] to a game you haven't played yet.

1. https://arxiv.org/abs/1803.03635

Re: Map of an Insect’s Brain

#54

Post by lossolo »

FrustratedMonky wrote: post #6

if you map every single neuron, at what point is it no longer just a 'model' of reality, and is reality. How do we know that the computer model of neurons does not have the same internal awareness as the real neurons?

> if you map every single neuron, at what point is it no longer just a 'model' of reality, and is reality

If you map position of every fish in the sea will you have a model of a living sea with living fish and all their interactions (electrical/physics/chemical etc)? Or just the positions of fish in a sea at certain point of time?

Re: Map of an Insect’s Brain

#55

Post by bombolo »

neurohdmi wrote: post #52
yyyk wrote: post #8

Earlier quoted context omitted.

>Is it simulated to the level where you could give it some visual stimulus and observe the actions it is trying to take? No way. First, this mapping doesn't tell us how the synapses are regulated - if we could 'run' this the weights would stay fixed forever and that's not how a brain works. Second, there must be some neurons dedicated to chemical management, and they'd go haywire unless you found a way to deal with t…

The first thing you said is correct. To do a proper simulation you would need to gather functional properties of the various cell classes and their synaptic connection, which this study didn't do. (Maybe you can find that information from other lab, I'm not familiar with fly models?) However we definitely have the computational ability to do simulations a fly network. Look at some of the modeling done by the Blue Bra…

But, do we know that they actually model completely accurately?

Re: Map of an Insect’s Brain

#56

Post by epgui »

lossolo wrote: post #51
xyproto wrote: post #31

Earlier quoted context omitted.

It could be simulated slowly at first.

To simulate something you need to first know how it works and know all the relevant interactions, this is something that we currently do not understand. Neurons are not equivalent of neural networks we use in computers, they are A LOT more complex with whole groups firing the same time and chemicals regulating neurons all the time and whole topology is plastic, it works in a way we can't model or simulate currently.…

I'm a biochemist, and I somewhat disagree with this take.

Yes, it's true that modelling just a single cell's interactions with its environments is beyond our capacity.

But here we're talking about simulating how a brain reacts to signals at a higher level of abstraction: we're studying an "emergent" phenomenon. We don't need to model molecular interactions, and we absolutely can model this using artificial neural networks.

This doesn't mean we'll be able to completely accurately model its behaviour, but we should get a lot closer than many HN commenters seem to believe. Biological neurons don't have magical properties, they just have more side-effects.

Re: Map of an Insect’s Brain

#57

Post by pmoriarty »

usgroup wrote: post #4

i intuitively think that there is something similar in kind between a fruit fly brain and a human brain regardless of the scale, and that cracking the smaller case will somehow lead to cracking the larger one. However, rather than hope, it leads me to a negative conclusion. That is, even at orders of magnitude smaller scales, and with full knowledge of how all the gross bits are connected, we still won't be able to u…

"the old microscopic will become the new macroscopic and we will realise that there is yet orders of magnitude more details to make sense of."

That's already the case.

Below the neuron level is the molecular level.

Below the molecular level is the atomic level.

Below the atomic level is the subatomic level.

The last of these is still in the process of being explored and understood by physics, and there it might not even be possible to measure all there is at that level, much less make sense of it or adequately model it.

As Roger Penrose famously pointed out, events at the subatomic level might be critical for consciousness, and we're very far from modeling even all the molecular interaction that happen in a human brain, nevermind the atomic or subatomic interactions.

Re: Map of an Insect’s Brain

#59

Post by epgui »

coldtea wrote: post #47
aperrien wrote: post #29

Earlier quoted context omitted.

I think you may be overestimating the complexity. The better idea is to set up and experiment with different simulation parameters, and see how far they diverge from actual observed behavior.

I think you underestimating the problem. There's no way to even capture and record "observed behavior" at the required level atm...

I'm a biochemist and I think you're under-estimating how sophisticated fruit fly behaviour studies are.

Re: Map of an Insect’s Brain

#60

Post by pmoriarty »

DonHopkins wrote: post #36
jokoon wrote: post #28

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

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.

Post reply on HN