Corpus ID: 36663713

Variational Autoencoder based Anomaly Detection using Reconstruction Probability

@inproceedings{An2015VariationalAB,
  title={Variational Autoencoder based Anomaly Detection using Reconstruction Probability},
  author={Jinwon An and S. Cho},
  year={2015}
}
  • Jinwon An, S. Cho
  • Published 2015
  • Computer Science
  • We propose an anomaly detection method using the reconstruction probability from the variational autoencoder. The reconstruction probability is a probabilistic measure that takes into account the variability of the distribution of variables. The reconstruction probability has a theoretical background making it a more principled and objective anomaly score than the reconstruction error, which is used by autoencoder and principal components based anomaly detection methods. Experimental results… CONTINUE READING

    Figures, Tables, and Topics from this paper.

    Interpreting Rate-Distortion of Variational Autoencoder and Using Model Uncertainty for Anomaly Detection
    Estimation of Dimensions Contributing to Detected Anomalies with Variational Autoencoders
    3
    Inverse-Transform AutoEncoder for Anomaly Detection
    11
    MAL DATA MANIFOLD FOR ANOMALY LOCALIZATION
    Anomaly Detection with Conditional Variational Autoencoders
    1
    Anomaly localization by modeling perceptual features
    Continual Learning for Anomaly Detection with Variational Autoencoder
    4
    A Two-class Hyper-spherical Autoencoder for Supervised Anomaly Detection
    6

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 15 REFERENCES
    Higher Order Contractive Auto-Encoder
    172
    Anomaly detection: A survey
    6477
    Contractive Auto-Encoders: Explicit Invariance During Feature Extraction
    991
    Variational Bayesian Inference with Stochastic Search
    294