Deep Learning without Poor Local Minima

Deep Learning without Poor Local Minima

In this paper, we prove a conjecture published in 1989 and also partially address an open problem announced at the Conference on Learning Theory (COLT) 2015. For an expected loss function of a deep nonlinear neural network, we prove the following statements under the independence assumption adopted...

Full description

Bibliographic Details
Main Author: Kawaguchi, Kenji
Other Authors: Leslie Kaelbling
Published: 2016
Subjects:
Optimization
Neural Network
Machine Learning
High Dimension
Convex
Non-convex
Local minimum
Global minimum
Saddle point
Critical point
Online Access:http://hdl.handle.net/1721.1/102665

Internet

http://hdl.handle.net/1721.1/102665

Similar Items