Awesome

Posterior Network: Uncertainty Estimation without OOD Samples via Density-Based Pseudo-Counts

This repository presents the experiments of the paper:

Posterior Network: Uncertainty Estimation without OOD Samples via Density-Based Pseudo-Counts<br> Bertrand Charpentier, Daniel Zügner, Stephan Günnemann<br> Conference on Neural Information Processing Systems (NeurIPS), 2020.

[Paper|Video]

Requirements

To install requirements:

conda env create -f environment.yaml
conda activate posterior-network
conda env list

Training & Evaluation

To train the model(s) in the paper, run one jupyter notebook in the folder notebooks. All parameter are described.

To dowload the datasets, follow the following links:

• 2DGaussians vs anomalous2D
• Segment (No sky) vs Segment (Sky only)
• SensorlessDrive (No 10, 11) vs SensorlessDrive (10, 11 only)
• MNIST vs FashionMNIST / KMNIST
• CIFAR10 vs SVHN

Pre-trained Models

You can find pre-trained models in the folder saved_models. Models in saved_models/MNIST-postnet and saved_models/CIFAR10-postnet are trained on classic MNIST and CIFAR10 splits.

Cite

Please cite our paper if you use the model or this code in your own work:

@incollection{postnet,
title = {Posterior Network: Uncertainty Estimation without OOD Samples via Density-Based Pseudo-Counts},
author = {Charpentier, Bertrand, Daniel Z\"{u}gner and G\"{u}nnemann, Stephan},
booktitle = {Advances in Neural Information Processing Systems 33},
year = {2020},
publisher = {Curran Associates, Inc.},
}