发表机构
KAIST; Pusan National University（韩国科学技术院; 釜山国立大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对现有多元时间序列异常检测方法忽略随机性且难以捕捉长时依赖和变量交互的问题，提出GT-PSSM，统一概率框架结合PSSM与图变换器，实现更稳健的异常检测。
AI 中文摘要
多元时间序列异常检测（MTAD）对于确保复杂系统安全可靠运行至关重要。许多现有方法通过在主要正常数据上训练重建或预测模型来学习正常模式。然而，这些方法中有很大一部分依赖确定性模型及其相关的逐点输出误差来进行异常评分。由于真实世界的多元时间序列因测量噪声和系统内在随机性而固有地具有随机性，纯基于误差的评分可能不可靠，因为大误差可能源于良性波动而非真正的异常。概率方法通过量化模型输出的不确定性来解决这一局限性。特别是，概率状态空间模型（PSSMs）通过潜在状态转移建模随机系统动力学，并通过发射模型建模测量噪声，提供了一个原则性框架。尽管有这一优势，现有的基于PSSM的MTAD方法通常难以捕捉长时程时间依赖和变量间依赖，因为它们通常依赖对噪声敏感的循环架构，并且缺乏显式的跨变量结构建模。为解决这些局限性，我们提出了图变换器增强的概率状态空间模型（GT-PSSM），这是一种新颖的基于PSSM的MTAD方法，它将基于PSSM的随机动力学概率建模与基于图变换器的时间及变量间依赖学习紧密结合。通过在统一概率框架中联合建模随机性、长时程时间依赖和变量交互，GT-PSSM实现了更稳健的异常检测。
英文摘要
Multivariate time series anomaly detection (MTAD) is crucial for ensuring the safe and reliable operation of complex systems. Many existing methods learn normal patterns by training reconstruction or forecasting models on predominantly normal data. However, a large portion of these approaches rely on deterministic models and their associated point-wise output errors for anomaly scoring. Since real-world multivariate time series are inherently stochastic due to measurement noise and intrinsic system randomness, purely error-based scores can be unreliable, as large errors may arise from benign fluctuations rather than true anomalies. Probabilistic approaches address this limitation by quantifying uncertainty in model outputs. In particular, probabilistic state-space models (PSSMs) provide a principled framework by modeling stochastic system dynamics through latent state transitions and measurement noise via emission models. Despite this advantage, existing PSSM-based MTAD methods often struggle to capture long-range temporal dependencies and inter-variable dependencies, as they typically rely on noise-sensitive recurrent architectures and lack explicit cross-variable structure modeling. To address these limitations, we propose Graph-Transformer-Enhanced Probabilistic State-Space Model (GT-PSSM), a novel PSSM-based MTAD method that tightly integrates PSSM-based probabilistic modeling of stochastic dynamics with Graph Transformer-based learning of temporal and inter-variable dependencies. By jointly modeling stochasticity, long-range temporal dependence, and variable interactions within a unified probabilistic framework, GT-PSSM enables more robust anomaly detection.
