Adaptive Variational Autoencoders for Outlier Detection in Data Streams: Unterschied zwischen den Versionen

Aus IPD-Institutsseminar
Zur Navigation springen Zur Suche springen
Zeile 11: Zeile 11:
  
 
Adaptive Variational Autoencoders instantiate the AGN framework with a Variational Autoencoder as generative model and automatically adapt the model to an occurring concept drift.
 
Adaptive Variational Autoencoders instantiate the AGN framework with a Variational Autoencoder as generative model and automatically adapt the model to an occurring concept drift.
AVAs generate their network architecture based on the dimensionality of incoming data.
+
AVA generates its network architecture based on the dimensionality of incoming data.
  
 
Our experiments against several benchmark outlier data sets show that AVA outperforms the state of the art and successfully adapts to streams with concept drift.
 
Our experiments against several benchmark outlier data sets show that AVA outperforms the state of the art and successfully adapts to streams with concept drift.
 
}}
 
}}

Version vom 25. März 2019, 21:05 Uhr

Vortragende(r) Florian Pieper
Vortragstyp Masterarbeit
Betreuer(in) Edouard Fouché
Termin Fr 29. März 2019
Vortragsmodus
Kurzfassung Outlier detection targets at the discovery of abnormal data patterns. Adaptive Variational Autoencoders (AVA) are a novel approach for unsupervised outlier detection in data streams.

We present a streaming framework for training arbitrary generative models such as Variational Autoencoders (VAE) in data streams, which we name Adaptive Generative Networks (AGN). The unique property of a generative model is its ability of generating samples from the model.

Adaptive Variational Autoencoders instantiate the AGN framework with a Variational Autoencoder as generative model and automatically adapt the model to an occurring concept drift. AVA generates its network architecture based on the dimensionality of incoming data.

Our experiments against several benchmark outlier data sets show that AVA outperforms the state of the art and successfully adapts to streams with concept drift.