Hanson.HSChang

Two plots show critical-point branches and their costs as regularization increases. Solid lines denote minima, dashed lines denote saddle points, and the principal branch has the lowest cost among the plotted solutions.

Review

How regularization affects the critical points in linear neural networks

November 08, 2018

The existence and optimality properties of the critical points of the linear neural networks with mean-squared loss function in the face of regularization.