3 ms·
> Internally the weights evolve according to solutions of differential equations. Is there a closed form representation for gradient descent?
by KRAKRISMOTT 4y ago
> Internally the weights evolve according to solutions of differential equations.
Is there a closed form representation for gradient descent?