Batch uniformization trains autoencoders for sound anomaly detection by minimizing a density-weighted average of anomaly scores estimated via kernel density estimation on mini-batches to achieve uniform scores for normal sounds.
The weighted average of the anomaly score was minimized, and the weight was defined as the reciprocal of the probabilistic density of each sam- ple
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Batch Uniformization for Minimizing Maximum Anomaly Score of DNN-based Anomaly Detection in Sounds
Batch uniformization trains autoencoders for sound anomaly detection by minimizing a density-weighted average of anomaly scores estimated via kernel density estimation on mini-batches to achieve uniform scores for normal sounds.