首页> 外文会议>20th European conference on artificial intelligence >Adversarial Label Flips Attack on Support Vector Machines
【24h】

Adversarial Label Flips Attack on Support Vector Machines

机译:支持向量机的对抗性标签翻转攻击

获取原文
获取原文并翻译 | 示例

摘要

To develop a robust classification algorithm in the adversarial setting, it is important to understand the adversary's strategy. We address the problem of label flips attack where an adversary contaminates the training set through flipping labels. By analyzing the objective of the adversary, we formulate an optimization framework for finding the label flips that maximize the classification error. An algorithm for attacking support vector machines is derived. Experiments demonstrate that the accuracy of classifiers is significantly degraded under the attack.
机译:为了在对抗环境中开发出鲁棒的分类算法,了解对手的策略很重要。我们解决了标签翻转攻击的问题,其中对手通过翻转标签污染了训练集。通过分析对手的目标,我们制定了一个优化框架,以查找使分类错误最大化的标签翻转。推导了一种攻击支持向量机的算法。实验表明,分类器的准确性在攻击下明显降低。

著录项

  • 来源
  • 会议地点 Montpellier(FR)
  • 作者单位

    Institute of Informatics, Technische Universitaet Muenchen, Germany;

    Institute of Informatics, Technische Universitaet Muenchen, Germany;

    Institute of Informatics, Technische Universitaet Muenchen, Germany;

  • 会议组织
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

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