Analytical solution for min_{c} |cx_1 + (1-c)x_2|_2^2 d is the distance (objective) optimzed v1v1 = <x1,x1> v1v2 = <x

Find the minimum norm solution as combination of two points This is correct only in 2D ie. min_c |\sum c_i x_i|_2^2 st. \sum

for reproducibility :param seed: :return:

Given y, it solves argmin_z |y-z|_2 st \sum z = 1 , 1 >= z_i >= 0 for all i

Parameters ---------- losses : shared_parameters : task_specific_parameters : last_shared_parameters

Given a list of vectors (vecs), this method finds the minimum norm element in the convex hull as min |u|_2 st. u = \sum c_i vecs[i] a

Parameters ---------- losses : shared_parameters : shared parameters kwargs : Returns -----

Parameters ---------- losses : shared_parameters : shared parameters kwargs : Returns -------

Parameters of the last shared layer. Returns -------

Adaptation of code from: https://github.com/Cranial-XIX/CAGrad

Parameters ---------- n_tasks : iteration_window : 'iteration' loss is averaged over the last 'iteration_window' los

:param method:

Given a list of vectors (vecs), this method finds the minimum norm element in the convex hull as min |u|_2 st. u = \sum c_i vecs[i] a

Parameters ---------- losses : shared_parameters : last_shared_parameters : representation :