ATSDLN uses FCN representation learning plus detector and parameter selection sub-networks to adaptively choose anomaly detectors for time series, claiming better performance and expandability via transfer learning on public datasets.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
An Adaptive Approach for Anomaly Detector Selection and Fine-Tuning in Time Series
ATSDLN uses FCN representation learning plus detector and parameter selection sub-networks to adaptively choose anomaly detectors for time series, claiming better performance and expandability via transfer learning on public datasets.