BriefGPT.xyz
Nov, 2017
无监督深度域自适应的最小熵相关对齐
Minimal-Entropy Correlation Alignment for Unsupervised Deep Domain Adaptation
HTML
PDF
Pietro Morerio, Jacopo Cavazza, Vittorio Murino
TL;DR
采用深度学习方法实现无监督域自适应，通过研究源域和目标域之间的二阶矩对齐可以最小化熵，提出了一种比欧几里得方法更有原则性的策略，利用源到目标的正则化器以无监督和数据驱动的方式量化，并在标准性的域适应基准测试中证实了该框架的优势。
Abstract
In this work, we face the problem of
unsupervised domain adaptation
with a novel
deep learning
approach which leverages on our finding that
entro
→