Explainable AI for microseismic event detection
Ayrat Abdullin, Denis Anikiev and Umair bin Waheed
像PhaseNet这样的深度神经网络在检测微地震事件方面表现出高精度,但它们的黑箱性质在关键应用中是一个值得关注的问题。 我们应用可解释的AI(XAI)技术,如梯度加权类激活映射(Grad-CAM)和Shapley Additive Explanations(SHAP),来解释PhaseNet模型的决策并提高其可靠性。 Grad-CAM强调,该网络的注意力与P波和S波到达一致。 SHAP值量化特征贡献,确认垂直组分幅度驱动P相选择,而水平组件主导S相位选择,与地球物理原理一致。 利用这些见解,我们引入了一个SHAP门控推理方案,将模型的输出与基于解释的指标相结合,以减少错误。 在9000波形的测试中,SHAP门控模型实现了F1的0.98分(精度0.99,召回0.97),优于基线PhaseNet(F1-score 0.97),并显示出增强的噪声稳健性。 这些结果表明,XAI不仅可以解释深度学习模型,还可以直接增强其性能,为建立自动化地震探测器的信任提供了一个模板。
Deep neural networks like PhaseNet show high accuracy in detecting microseismic events, but their black-box nature is a concern in critical applications. We apply explainable AI (XAI) techniques, such as Gradient-weighted Class Activation Mapping (Grad-CAM) and Shapley Additive Explanations (SHAP), to interpret the PhaseNet model's decisions and improve its reliability. Grad-CAM highlights that the network's attention aligns with P- and S-wave arrivals. SHAP values quantify feature contributions...