RedLamp detects time series anomalies by training a multiclass classifier on 11 types of augmented pseudo-anomalies, combining its predictions with reconstruction error, and using soft labels to resist contamination.
Carmona, François-Xavier Aubet, Valentin Flunkert, and Jan Gasthaus
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Robust and Explainable Detector of Time Series Anomaly via Augmenting Multiclass Pseudo-Anomalies
RedLamp detects time series anomalies by training a multiclass classifier on 11 types of augmented pseudo-anomalies, combining its predictions with reconstruction error, and using soft labels to resist contamination.