This repository contains the code for the paper Implicit Weight Uncertainty in Neural Networks (arXiv).
Modern neural networks tend to be overconfident on unseen, noisy or incorrectly labelled data and do not produce meaningful uncertainty measures. Bayesian deep learning aims to address this shortcoming with variational approximations (such as Bayes by Backprop or Multiplicative Normalising Flows). However, current approaches have limitations regarding flexibility and scalability. We introduce Bayes by Hypernet (BbH), a new method of variational approximation that interprets hypernetworks as implicit distributions. It naturally uses neural networks to model arbitrarily complex distributions and scales to modern deep learning architectures. In our experiments, we demonstrate that our method achieves competitive accuracies and predictive uncertainties on MNIST and a CIFAR5 task, while being the most robust against adversarial attacks.
Following libraries were used for development:
future==0.16.0
jupyter==1.0.0
matplotlib==2.2.2
notebook==5.0.0
numpy==1.14.3
observations==0.1.4
pandas==0.19.2
scikit-learn==0.19.1
scipy==1.1.0
seaborn==0.8.1
tensorflow-gpu==1.7.0
tqdm==4.19.5
toy_data.ipynb contains the code for the toy regression.
The other files contain the code for the mnist and cifar experiments.
run_* just calls the experiments. base_layers and layers
implement easy to use layers for different VI methods. networks
holds the models and the actual training and evaluation is in
experiments and utils.
For discussion, suggestions or questions don't hesitate to contact [email protected] .