This is the source code of the AISTATS paper
Emilio Dorigatti, Benjamin Schubert, Bernd Bischl, David Ruegamer, Frequentist Uncertainty Quantification in Semi-Structured Neural Networks, Proceedings of The 26th International Conference on Artificial Intelligence and Statistics, PMLR 206:1924-1941, 2023.
https://proceedings.mlr.press/v206/dorigatti23a.html
The folder theory-simulations
contains the code to reproduce the simulations and figures 2-6 of the paper.
To generate figures 4-6, first execute poisson_sddr_run.R
to train the neural networks, then the respective script for the figure.
The folder skin-lesion
contains the code to reproduce the practical application on the skin lesion dataset:
- Download the dataset into the
skin-lesion/data
folder: https://challenge2020.isic-archive.com/- Choose
Download metadata v2 (2MB)
- Choose
Download DICOM Corrected* (23.0GB)
and unzip the images intodata/train
- Choose
- Run the first data-preparation script:
Rscript preprocess-1.R
- Run the second data-preparation script:
python preprocess-2.py
- Run cross-validation and grid search:
bash train.sh
- Aggregate the predictions
python run.py aggregate
- Fit GAMM models on the networks:
bash fit-gamm.sh
- Finally, analyze their predictions:
Rscript analyze-gamm.R
The theory simulations with the neural networks use the R interface to Keras, while the skin lesion example uses pytorch and pytorch-lightning.