Aktivität: Vortrag ohne Tagungsband / Vorlesung › Präsentation auf einer wissenschaftlichen Konferenz / Workshop
We applied convolutional versions of a standard autoencoder (CAE), a variational autoencoder (VAE) and an adversarial autoencoder (AAE) to two different publicly available datasets and compared their anomaly detection performances in a qualitative and quantitative manner. The time needed for training the models is measured to capture their complexity. The simplest model, the CAE, computes results which are nearly as accurate and for some cases even better than results achieved by the VAE and AAE. All three autoencoder types computed convincing anomaly detection results for the more simple-structured MNIST scenario. However, none of the autoencoder types proved to capture a good representation of the relevant features of the more complex CIFAR10 dataset, leading to moderately good anomaly detection performances.