FocusCOMPUTER VISION

Toward principled regularization of deep networks—From weight decay to feature contraction

See allHide authors and affiliations

Science Robotics  15 May 2019:
Vol. 4, Issue 30, eaaw1329
DOI: 10.1126/scirobotics.aaw1329

You are currently viewing the abstract.

View Full Text

Log in to view the full text

Log in through your institution

Log in through your institution

Abstract

Training deep artificial neural networks for classification problems may benefit from exploiting intrinsic class similarities by way of network regularization that compensates for a drawback in the commonly used target error.

View Full Text