|
非对称风险最小化度量异常检测
Metric Anomaly Detection Via Asymmetric Risk Minimization |
|
| |
| 课程网址: |
http://videolectures.net/simbad2011_menahem_minimization/
|
| 主讲教师: |
Eitan Menahem
|
| 开课单位: |
内盖夫本-古里安大学
|
| 开课时间: |
2011-10-17
|
| 课程语种: |
英语
|
| 中文简介: |
我们提出了似乎是第一个只从正面例子中学习的异常检测框架，并且对正常点和异常点的表现和惩罚的实质性差异非常敏感。我们的框架引入了一种新型的不对称性，即错误报警（将正常实例误分类为异常）和遗漏异常（将异常误分类为正常）是如何受到惩罚的：尽管每个错误报警都会产生单位成本，但我们的模型假定，如果遗漏一个或多个异常，则会产生较高的全局成本。我们定义了一些自然的风险概念以及有效的最小化算法。我们的框架适用于任何有限倍维的度量空间。我们做了最低限度的假设，自然地概括了欧几里得空间中的边际等概念。我们对风险进行了理论分析，表明在温和条件下，我们的分类器是渐近一致的。我们提出的学习算法在计算和统计上是有效的，并且在运行时间和精度之间存在进一步的权衡。给出了一些实际数据的实验结果。
|
| 课程简介: |
We propose what appears to be the first anomaly detection framework that learns from positive examples only and is sensitive to substantial differences in the presentation and penalization of normal vs. anomalous points. Our framework introduces a novel type of asymmetry between how false alarms (misclassifications of a normal instance as an anomaly) and missed anomalies (misclassifications of an anomaly as normal) are penalized: whereas each false alarm incurs a unit cost, our model assumes that a high global cost is incurred if one or more anomalies are missed. We define a few natural notions of risk along with efficient minimization algorithms. Our framework is applicable to any metric space with a finite doubling dimension. We make minimalistic assumptions that naturally generalize notions such as margin in Euclidean spaces. We provide a theoretical analysis of the risk and show that under mild conditions, our classifier is asymptotically consistent. The learning algorithms we propose are computationally and statistically efficient and admit a further tradeoff between running time and precision. Some experimental results on real-world data are provided.
|
| 关 键 词: |
计算机科学; 机器学习; 度量空间
|
| 课程来源: |
视频讲座网
|
| 最后编审: |
2020-06-03：魏雪琼(课程编辑志愿者)
|
| 阅读次数: |
44
|
|
图 书 馆