You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We tried normalizing by the size of the summed area (effectively making it an adaptive average pooling) but this normalization made the training unstable. We believe this is because we optimize the parameters ᾶ^l and ᾶ^r for both the optimal window size as well as bringing the magnitude of the representation down.
Is it possible to divide the output of TaLK to the area of span?
area = (left_offset + right_offset) * kernel
output = integral / area
The text was updated successfully, but these errors were encountered: