DriftSurf: Stable-State / Reactive-State Learning under Concept Drift
DriftSurf: Stable-State / Reactive-State Learning under Concept Drift
复制标题
DOI：
--
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Ashraf Tahmasbi;Ellango Jothimurugesan;Srikanta Tirthapura;Phillip B. Gibbons
中科院分区：
文献类型：
--
作者：
Ashraf Tahmasbi;Ellango Jothimurugesan;Srikanta Tirthapura;Phillip B. Gibbons
当从流数据中学习时，数据分布的变化，也称为概念漂移，可能会使先前学习的模型不准确，需要训练新的模型。我们提出了一种自适应学习算法，通过将漂移检测结合到更广泛的稳态/重激活过程中，扩展了以前基于漂移检测的方法。该方法的优点是，我们可以在稳定状态下使用主动漂移检测来获得高的检测率，但通过对真实漂移做出快速反应的反应状态来降低独立漂移检测的误检率，同时消除大多数误报。该算法在其基本学习器中是通用的，并且可以应用于各种监督学习问题。我们的理论分析表明，该算法的风险是(I)在统计上优于独立的漂移检测，以及(Ii)与先知知道何时发生(突然)漂移的算法相竞争。在具有概念漂移的合成数据集和真实数据集上的实验与我们的理论分析一致。
When learning from streaming data, a change in the data distribution, also known as concept drift, can render a previously-learned model inaccurate and require training a new model. We present an adaptive learning algorithm that extends previous drift-detection-based methods by incorporating drift detection into a broader stable-state/reactivestate process. The advantage of our approach is that we can use aggressive drift detection in the stable state to achieve a high detection rate, but mitigate the false positive rate of standalone drift detection via a reactive state that reacts quickly to true drifts while eliminating most false positives. The algorithm is generic in its base learner and can be applied across a variety of supervised learning problems. Our theoretical analysis shows that the risk of the algorithm is (i) statistically better than standalone drift detection and (ii) competitive to an algorithm with oracle knowledge of when (abrupt) drifts occur. Experiments on synthetic and real datasets with concept drifts confrm our theoretical analysis.