3 ms·

I haven't read the paper itself, but the article makes it sound like they simply counted the minimum number of bits required to represent the English language.

I haven't read the paper itself, but the article makes it sound like they simply counted the minimum number of bits required to represent the English language. It mentions no neuroscience-specific insights, so the "brain" part of the title is quite misleading.

rob74 8y ago
Aside from that - is there really a way to express neural connections in the brain in terms of bits and bytes?
- jerf 8y ago
  The problem isn't whether there is "a" way; the problem is there's a multiplicity of ways and we don't know what the best one is. Part of the reason we don't have any idea what the best is is that we can't even construct any of them from a real brain, that is, the "brain scanning" technology the techno-rapture-style Singularity expects to develop does not even remotely exist right now. So we have nothing to experiment with or gain any experience with. The only ones we'd even remotely have access to right now would be some rather brute-force encoding of neurons and connectivity strengths and we pretty much know that such a thing must be incredibly redundant, but we have no ability right now to remove that redundancy. And we can't do it on a very large neural set. It has been done on small models, though, at least to some significant extent: https://en.wikipedia.org/wiki/OpenWorm https://en.wikipedia.org/wiki/OpenWorm
  - aperrien 8y ago
    Brain scanning techniques and simulation software have a long way to go, but there have been significant advances made: A full fly connectome: https://www.nature.com/articles/s41684-018-0183-8 https://www.nature.com/articles/s41684-018-0183-8 Architecture of the Mouse Brain Synaptome: https://www.cell.com/neuron/fulltext/S0896-6273(18)30581-6 https://www.cell.com/neuron/fulltext/S0896-6273(18)30581-6 So we are past just a simple worm simulation now.
    - jerf 8y ago
      Thank you for the better links. I just meant that as an example.
- jtloong 8y ago
  No. We still have a poor understanding of the exact way neurons code information and store information. This page shows some of the current theories: https://en.wikipedia.org/wiki/Neural_coding#Hypothesized_coding_schemes https://en.wikipedia.org/wiki/Neural_coding#Hypothesized_cod...
- randyrand 8y ago
  most definitely. ANN's do just that. That said, they assume a computer. and a program to run it. Similarly, the brain probably has many complex lower structures that compress language storage, but the compression software is not trivial to store itself.

rob74 8y ago

Aside from that - is there really a way to express neural connections in the brain in terms of bits and bytes?

jerf 8y ago
The problem isn't whether there is "a" way; the problem is there's a multiplicity of ways and we don't know what the best one is. Part of the reason we don't have any idea what the best is is that we can't even construct any of them from a real brain, that is, the "brain scanning" technology the techno-rapture-style Singularity expects to develop does not even remotely exist right now. So we have nothing to experiment with or gain any experience with. The only ones we'd even remotely have access to right now would be some rather brute-force encoding of neurons and connectivity strengths and we pretty much know that such a thing must be incredibly redundant, but we have no ability right now to remove that redundancy. And we can't do it on a very large neural set. It has been done on small models, though, at least to some significant extent: https://en.wikipedia.org/wiki/OpenWorm https://en.wikipedia.org/wiki/OpenWorm
- aperrien 8y ago
  Brain scanning techniques and simulation software have a long way to go, but there have been significant advances made: A full fly connectome: https://www.nature.com/articles/s41684-018-0183-8 https://www.nature.com/articles/s41684-018-0183-8 Architecture of the Mouse Brain Synaptome: https://www.cell.com/neuron/fulltext/S0896-6273(18)30581-6 https://www.cell.com/neuron/fulltext/S0896-6273(18)30581-6 So we are past just a simple worm simulation now.
  - jerf 8y ago
    Thank you for the better links. I just meant that as an example.
jtloong 8y ago
No. We still have a poor understanding of the exact way neurons code information and store information. This page shows some of the current theories: https://en.wikipedia.org/wiki/Neural_coding#Hypothesized_coding_schemes https://en.wikipedia.org/wiki/Neural_coding#Hypothesized_cod...
randyrand 8y ago
most definitely. ANN's do just that. That said, they assume a computer. and a program to run it. Similarly, the brain probably has many complex lower structures that compress language storage, but the compression software is not trivial to store itself.