Empirical study on tangent loss function for classification with deep neural networks

Published: 01 Jan 2021, Last Modified: 25 Jul 2025Comput. Electr. Eng. 2021EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•Aiming to make classification models more stable.•Making a systematical empirical study on tangent loss.•Comparing with cross-entropy loss on various classification tasks.•Achieving better or competitive performance than cross-entropy loss.
Loading