[论文解读] Variational Autoencoders for P-wave Detection on Strong Motion Earthquake Spectrograms

Türkan Simge İşpak, Salih Tileylioğlu+1 more|arXiv (Cornell University)|Jan 9, 2026

Seismology and Earthquake Studies被引用 0

一句话总结

该论文系统比较四种VAE架构（Basic、Skip、Attention、Hybrid），研究在强烈运动频谱图中P波检测的重建保真度与异常检测之间的权衡对检测的影响，发现基于注意力的VAE在检测方面表现最佳。

ABSTRACT

Accurate P-wave detection is critical for earthquake early warning, yet strong-motion records pose challenges due to high noise levels, limited labeled data, and complex waveform characteristics. This study reframes P-wave arrival detection as a self-supervised anomaly detection task to evaluate how architectural variations regulate the trade-off between reconstruction fidelity and anomaly discrimination. Through a comprehensive grid search of 492 Variational Autoencoder configurations, we show that while skip connections minimize reconstruction error (Mean Absolute Error approximately 0.0012), they induce "overgeneralization", allowing the model to reconstruct noise and masking the detection signal. In contrast, attention mechanisms prioritize global context over local detail and yield the highest detection performance with an area-under-the-curve of 0.875. The attention-based Variational Autoencoder achieves an area-under-the-curve of 0.91 in the 0 to 40-kilometer near-source range, demonstrating high suitability for immediate early warning applications. These findings establish that architectural constraints favoring global context over pixel-perfect reconstruction are essential for robust, self-supervised P-wave detection.

研究动机与目标

• 研究VAE的结构偏置如何影响强动谱图上的P波检测中的重建与异常检测
• 评估四种VAE变体（Basic-VAE、Skip-VAE、Attention-VAE、Hybrid-VAE）在大规模配置网格上的表现
• 确定能在最大化检测性能（AUC）的同时管理重建质量（MAE）的架构配置
• 评估潜在容量和Transformer超参数如何影响近源到远源范围的鲁棒性与稳定性

提出的方法

• 将P波检测建模为在P波窗口上训练的VAE的自监督异常检测
• 通过预处理的强动加速度计记录生成包含伪影去除和基于FBM的噪声增强的声谱图输入
• 使用对称的编码器–解码器结构训练四种VAE变体（Basic-VAE、Skip-VAE、Attention-VAE、Hybrid-VAE）
• 使用MAE和NCC作为重建与相似度分数，并在滑动窗口上通过ROC-AUC评估检测
• 对潜在维度（32–256）进行网格搜索，对于注意力模型， transformers 深度（1–48）和头数，总计492组配置
• 通过最佳AUC及对应MAE比较架构，突出重建-检测权衡

Figure 1: Detection performance versus reconstruction quality for the four VAE architectures tested in our experiments. Detection is measured by the Area Under the Receiver Operating Characteristic Curve ( $\mathit{AUC}$ , higher is better), and reconstruction is measured by the Mean Absolute Error

实验结果

研究问题

• RQ1 哪些架构偏置（跳跃连接、自注意力）能够在强动谱图中的P波起始处实现重建保真度与异常检测之间的最佳平衡
• RQ2 潜在容量（32–256）和Transformer超参数如何影响近源距离下的检测稳定性与性能
• RQ3 在没有大量标注数据的情况下，注意力驱动的VAE是否能在自监督P波检测中超越以重建为重点的变体
• RQ4 与P波起始对齐的窗口相对对检测性能的影响在最佳架构中如何表现
• RQ5 源点到观测站距离如何影响所提模型的检测能力

主要发现

• Attention-VAE在所有架构中实现了最高的检测AUC（0.8749），但伴随较高的重建误差（MAE 0.00428）
• Skip-VAE的重建误差最低（MAE 0.00119），但检测性能较低（AUC 0.8210）
• Hybrid-VAE在AUC与MAE之间取得平衡，分别为0.8445与0.00213
• Basic-VAE总体性能最弱（AUC 0.8074，MAE 0.00461）
• 表现最佳的Attention-VAE在近源2 40 km内的AUC接近0.91，显示其对即时EEW的强适用性，且性能对时间对齐敏感（窗口在P波前1 s开始时达到峰值）
• 模型性能呈现权衡：更高的全局上下文提高检测，但像素级重建可能削弱异常判别能力

Figure 2: Unified diagram of the proposed VAE architectures. Components marked with an asterisk (*) are optional, defining four configurations: Basic-VAE (baseline encoder–decoder), Skip-VAE (incorporating long-range skip connections), Attention-VAE (incorporating the self-attention bottleneck), and