On: Artificial Neural Nets Grow Brainlike Navigation C...

[stochastic_monk]

The big difference is that biological neurons emit binary outputs. Because of this, there’s no gradient and they can’t be trained by SGD.

[yorwba → stochastic_monk]

But biological neurons are also stochastic, and the firing probabilities are continuous, so you can do SGD on them.

[danharaj → stochastic_monk]

There's phase and frequency information in neural spike trains.