Ask HN: Why doesn't AI use dynamical neurons instead of static activations?

1 pointstonii1415mo ago7 comments

Why doesn’t AI use dynamical neurons instead of static activations even when it is known that biological neurons are dynamical systems?

7 comments

rini175mo ago

I guess it's impossible to simulate billions of dynamical neurons with current computer architecture, it is limited by memory bandwidth.

tonii141OP5mo ago

I agree, but maybe there is no need for billions of neurons to be simulated right away. Artificial neural networks were pretty small at the time.

rini175mo ago

and they did not do much

1 more reply

nialse5mo ago

Dynamical neurons are not used because they add enormous computational cost and instability for limited practical gain.

tonii141OP5mo ago

What do you mean by "they add instability"?

nialse5mo ago

Neural nets are unstable during training and dynamic weights amplifies the problem. Thus, the neural networks could end up in totally unusable states at inference time.

j / k navigate · click thread line to collapse

Ask HN: Why doesn't AI use dynamical neurons instead of static activations?

1 pointstonii1415mo ago7 comments

Why doesn’t AI use dynamical neurons instead of static activations even when it is known that biological neurons are dynamical systems?

7 comments

rini175mo ago

I guess it's impossible to simulate billions of dynamical neurons with current computer architecture, it is limited by memory bandwidth.

tonii141OP5mo ago

I agree, but maybe there is no need for billions of neurons to be simulated right away. Artificial neural networks were pretty small at the time.

rini175mo ago

and they did not do much

1 more reply

nialse5mo ago

Dynamical neurons are not used because they add enormous computational cost and instability for limited practical gain.

tonii141OP5mo ago

What do you mean by "they add instability"?

nialse5mo ago

Neural nets are unstable during training and dynamic weights amplifies the problem. Thus, the neural networks could end up in totally unusable states at inference time.

j / k navigate · click thread line to collapse