<?xml version="1.0" encoding="utf-8" standalone="yes" ?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>任务无关知识 | ViLab</title>
    <link>https://vilab.team/tag/%E4%BB%BB%E5%8A%A1%E6%97%A0%E5%85%B3%E7%9F%A5%E8%AF%86/</link>
      <atom:link href="https://vilab.team/tag/%E4%BB%BB%E5%8A%A1%E6%97%A0%E5%85%B3%E7%9F%A5%E8%AF%86/index.xml" rel="self" type="application/rss+xml" />
    <description>任务无关知识</description>
    <generator>Hugo Blox Builder (https://hugoblox.com)</generator><language>en-us</language><lastBuildDate>Fri, 01 Jan 2021 00:00:00 +0000</lastBuildDate>
    <image>
      <url>https://vilab.team/media/icon_hu2896232876136423579.png</url>
      <title>任务无关知识</title>
      <link>https://vilab.team/tag/%E4%BB%BB%E5%8A%A1%E6%97%A0%E5%85%B3%E7%9F%A5%E8%AF%86/</link>
    </image>
    
    <item>
      <title>Task-independent knowledge makes for transferable representations for generalized zero-shot learning</title>
      <link>https://vilab.team/publication/task-independent-knowledge-makes-for-transferable-representa/</link>
      <pubDate>Fri, 01 Jan 2021 00:00:00 +0000</pubDate>
      <guid>https://vilab.team/publication/task-independent-knowledge-makes-for-transferable-representa/</guid>
      <description>&lt;p&gt;本文针对广义零样本学习（GZSL）中可见类与未见类之间的表示偏差问题，提出利用任务无关知识来学习可迁移的视觉表示。通过在大规模辅助数据上预训练或引入外部知识，使模型捕获与类别标签无关的通用特征，从而提升对未见类别的识别能力。在多个基准数据集上验证了方法的有效性。&lt;/p&gt;
</description>
    </item>
    
  </channel>
</rss>
