2 ms·
> Accessibility of data has obvious implications on the pace of progress in using said data. That is only true if the data is likely to be the right data. For
by tsimionescu 6y ago
> Accessibility of data has obvious implications on the pace of progress in using said data.
That is only true if the data is likely to be the right data. For motor control, electrical signals seem to be pretty much it. But we have no clear proof that the same is true for higher thought processes - it could well be that chemical signals and processing inside each neuron itself are major components of that, so the data would be massively limited.
It would also be interesting to see how well such an approach would do when tried on a microprocessor - could we tell that a microprocessor is displaying "hello world" by sampling electrical signals from its transistors and trying to apply some kind of machine learning on that data?
- bshanks 6y agoYou might like this article titled "Could a Neuroscientist Understand a Microprocessor?": https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1005268 https://journals.plos.org/ploscompbiol/article?id=10.1371/jo...