Skip to Main content Skip to Navigation
Reports

Adversarial autoencoders for novelty detection

Valentin Leveau 1 Alexis Joly 1
1 ZENITH - Scientific Data Management
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier, CRISAM - Inria Sophia Antipolis - Méditerranée
Abstract : In this paper, we address the problem of novelty detection, i.e recognizing at test time if a data item comes from the training data distribution or not. We focus on Adversarial autoencoders (AAE) that have the advantage to explicitly control the distribution of the known data in the feature space. We show that when they are trained in a (semi-)supervised way, they provide consistent novelty detection improvements compared to a classical autoencoder. We further improve their performance by introducing an explicit rejection class in the prior distribution coupled with random input images to the autoencoder.
Complete list of metadatas

Cited literature [10 references]  Display  Hide  Download

https://hal.inria.fr/hal-01636617
Contributor : Alexis Joly <>
Submitted on : Thursday, November 16, 2017 - 5:25:48 PM
Last modification on : Thursday, February 14, 2019 - 11:06:11 AM
Long-term archiving on: : Saturday, February 17, 2018 - 1:51:33 PM

File

valentin.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01636617, version 1

Collections

Citation

Valentin Leveau, Alexis Joly. Adversarial autoencoders for novelty detection. [Research Report] Inria - Sophia Antipolis. 2017. ⟨hal-01636617⟩

Share

Metrics

Record views

869

Files downloads

1288