4 ms·
" But it is reasonable to expect that they will get better, and it is reasonable to expect that we will be able to emulate the brain with transistors and on a m
by edblarney 10y ago
" But it is reasonable to expect that they will get better, and it is reasonable to expect that we will be able to emulate the brain with transistors and on a much bigger scale some days."
No, it's not reasonable to assume they will emulate brains.
Not even the experts are saying that.
They're not trying to emulate brains at all.
- cm2187 10y agoI was under the impression that neural networks are inspired by the brain. And as we understand how the brain works better, it is obvious that we will be replicating it mechanically.
- Ar-Curunir 10y agoOnly superficially. Deep learning has little to do with biology, and even if a future AGI was running on a NN, it's structure would be very different from our brains.
- otakucode 10y agoUhm... yes, they most definitely are. There are many efforts underway right now building 'neuromorphic' architectures with memristors (which function very similarly to neurons). I don't think there is any promise in the approach personally (our 'mind' is a property of the feedback loop between our body and the world, and can not exist without the body, dualism is just wrong) but there are many who disagree. Also, even going back to von Neumann it has always been expected that the brain will be able to be built in hardware, simply because neurons exhibit binary behavior. They fire or not, and all the rest is massive interconnectedness and adaptive changes to their activation potential. There's lots of argument about whether reproducing the behavior of neurons will result in a 'mind' or be useful, but no one would argue that we simply won't be able to build a brain workalike.