面向上下文的领域适应方法用于时序异常检测（英文版）

英文标题：Context-aware Domain Adaptation for Time Series Anomaly Detection中文摘要：通过深度增强学习和上下文采样来建立时间序列领域自适应模型，以实现相似或不同领域之间的知识转移，并且在三个公开数据集上取得了良好的效果。英文摘要：Time series anomaly detection is a challenging task with a wide range ofreal-world applications. Due to label sparsity, training a deep anomalydetector often relies on unsupervised approaches. Recent efforts have beendevoted to time series domain adaptation to leverage knowledge from similardomains. However, existing solutions may suffer from negative