Use η0 / (1 + λ η0 t)^0.75 by default instead of ...^2/3. #12
Description
The learning rate has the form η0 / (1 + λ η0 t)^0.75 where λ is the regularization constant.
{{ message }}
The learning rate has the form η0 / (1 + λ η0 t)^0.75 where λ is the regularization constant.