[PyTorch] Anomaly Detection using Convolutional Variational Auto-Encoder (CVAE)

Example of Anomaly Detection using Convolutional Variational Auto-Encoder (CVAE) [TensorFlow 1.x] [TensorFlow 2.x].

Architecture

Simplified VAE architecture.

Problem Definition

'Class-1' is defined as normal and the others are defined as abnormal.

Results

	MNIST	Fashion-MNIST
Reconstruciton of training
Latent of training
Latent walk
Latent of test
Histogram of test
AUROC	0.997	0.980

Environment

• Python 3.7.4
• PyTorch 1.1.0
• Numpy 1.17.1
• Matplotlib 3.1.1
• Scikit Learn (sklearn) 0.21.3

Reference

[1] Kingma, D. P., & Welling, M. (2013). Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114.
[2] Kullback Leibler divergence. Wikipedia