undefined | Better HN

0 pointsdietr1ch1y ago0 comments

I guess that if the bulk of the computation goes into the multiplications, you can work in the log-space and simply sum, and when the time comes to actually do a sum on the original space you can go back and sum.

0 comments

a-loup-e1y ago

Not sure how well that would work if you're often adding bias after every layer

j / k navigate · click thread line to collapse

0 pointsdietr1ch1y ago0 comments

I guess that if the bulk of the computation goes into the multiplications, you can work in the log-space and simply sum, and when the time comes to actually do a sum on the original space you can go back and sum.

0 comments

a-loup-e1y ago

Not sure how well that would work if you're often adding bias after every layer

j / k navigate · click thread line to collapse