arXiv：cs.LG· Aditi Gautam, Manikandan Ravikiran, Saira Qureshi, Sena Fatma Ekiz, Charan Jagwani, Mahmoud Elhadidy, Sanjeev Roka·· 13 小时前
NV-Tesseract-AD：课程学习增强的条件扩散用于多变量时间序列异常检测
NV-Tesseract-AD: Curriculum-Enhanced Conditional Diffusion for Multivariate Time-Series Anomaly Detection
AI 导读
NV-Tesseract-AD 提出一种课程学习增强的条件扩散框架，将重建难度沿掩码强度与扩散噪声水平两个轴从易到难结构化训练，F1 相比无课程基线提升约 3.7 倍、MSE 降低 56%。
正文
Abstract:We propose NV-Tesseract-AD, a curriculum learning-enhanced conditional diffusion framework for multivariate time-series anomaly detection. Our central hypothesis is that reconstruction difficulty can be explicitly structured along two complementary axes, masking severity and diffusion noise level, and that training from easier to harder tasks along both axes simultaneously stabilizes optimization and improves residual discriminability. We integrate a Transformer-based diffusion backbone with a three-stage dual curriculum over masking ratios and diffusion timesteps, retaining anomalous samples throughout training and relying on structured corruption to drive self-supervised representation learning. We evaluate NV-Tesseract-AD under two regimes: in-distribution training on the TSB-UAD-M corpus and strict zero-shot transfer to ten EasyTSAD benchmark families without per-dataset tuning. The dual-axis curriculum produces masking-invariant reconstruction, confirmed by a near-zero loss gap between complementary masking strategies within the first training stage, and yields an approximately 3.7-fold improvement in F1 and a 56 percent reduction in MSE relative to a non-curriculum baseline. Zero-shot performance reaches F1 = 0.984 on NAB and F1 = 0.948 on NormA. Weaker results on heterogeneous families are traced to unsupervised threshold calibration rather than reconstruction quality, identifying calibration as the dominant remaining detection bottleneck. Accelerated inference via DPM-Solver++ reduces sampling cost by a factor of 3.5 to 5.7 at the dataset level but couples to thresholding behavior, establishing that the sampler and threshold must be validated jointly. Together, these findings demonstrate that aligning diffusion denoising difficulty with curriculum learning principles provides a principled and practical approach for robust anomaly detection in complex multivariate time series.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.12754 [cs.LG] |
| (or arXiv:2610.12754v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.12754 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Aditi Gautam [view email]
[v1]
Thu, 8 Oct 2026 22:06:40 UTC (1,863 KB)
来源：arXiv：cs.LG · arxiv.org