Abstract: We propose an unsupervised open set recognition approach that uses an Adversarial Autoencoder (AAE) for clustering and combines it with ii-loss to learn a representation that facilitates ...

A PyTorch implementation of Adversarial Autoencoders (AAEs) for unsupervised classification ... Note: since the .ipynb file is quite large (lots of images!) it may not open correctly on GitHub. I ...

In real-world applications unknown images from outside these classes, called open-set images, can pollute the prediction task, leading to them being wrongly classified as closed-set images. Open-set ...