• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>IEEE/CVF Conference on Computer Vision and Pattern Recognition >Sequential Mastery of Multiple Visual Tasks: Networks Naturally Learn to Learn and Forget to Forget
【24h】

Sequential Mastery of Multiple Visual Tasks: Networks Naturally Learn to Learn and Forget to Forget

机译：顺序掌握多个视觉任务：网络自然地学会了学习而忘记了

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

We explore the behavior of a standard convolutional neural net in a continual-learning setting that introduces visual classification tasks sequentially and requires the net to master new tasks while preserving mastery of previously learned tasks. This setting corresponds to that which human learners face as they acquire domain expertise serially, for example, as an individual studies a textbook. Through simulations involving sequences of ten related visual tasks, we find reason for optimism that nets will scale well as they advance from having a single skill to becoming multi-skill domain experts. We observe two key phenomena. First, emph{forward facilitation}---the accelerated learning of task n+1 having learned n previous tasks---grows with n. Second, emph{backward interference}---the forgetting of the n previous tasks when learning task n+1 ---diminishes with n. Amplifying forward facilitation is the goal of research on metalearning, and attenuating backward interference is the goal of research on catastrophic forgetting. We find that both of these goals are attained simply through broader exposure to a domain.
机译：我们在连续学习的环境中探索标准卷积神经网络的行为，该环境顺序引入视觉分类任务，并要求网络掌握新任务，同时保持对先前学习任务的掌握。此设置对应于人类学习者在连续获得领域专业知识时所面对的环境，例如，当个人学习教科书时。通过涉及十个相关视觉任务序列的模拟，我们找到了乐观的理由，即随着网络从具有单一技能发展为多技能领域专家，网络将能够很好地扩展。我们观察到两个关键现象。首先，\ emph {forward facilitation}-已经学习了n个先前任务的对任务n + 1的加速学习-与n一起增长。其次，\ emph {向后干扰} ---学习任务n + 1时忘记了前n个任务-用n减少。扩大向前的促进作用是金属学习的目标，而减弱向后的干扰则是灾难性遗忘研究的目标。我们发现，这两个目标都是通过更广泛地接触某个领域来实现的。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号