Shiyu Liang, Ruoyu Sun, Jason D. Lee, R. Srikant
Published 2018-05-22Version 1
One of the main difficulties in analyzing neural networks is the non-convexity of the loss function which may have many bad local minima. In this paper, we study the landscape of neural networks for binary classification tasks. Under mild assumptions, we prove that after adding one special neuron with a skip connection to the output, or one special neuron per layer, every local minimum is a global minimum.
The layer-wise L1 Loss Landscape of Neural Nets is more complex around local minima
Eliminating all bad Local Minima from Loss Landscapes without even adding an Extra Unit
Deep Learning without Poor Local Minima
![QR code for arXiv:1805.08671 [stat.ML]](http://oss.arxitics.com/images/favicon.svg)