A conditional diffusion model with quartile normalization and Huber loss assigns outlier probabilities to univariate structural monitoring data points and produces a global quality score, outperforming clustering, isolation, and reconstruction baselines on real structures.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Probabilistic data quality assessment for structural monitoring data via outlier-resistant conditional diffusion model
A conditional diffusion model with quartile normalization and Huber loss assigns outlier probabilities to univariate structural monitoring data points and produces a global quality score, outperforming clustering, isolation, and reconstruction baselines on real structures.