Code to reproduce experiments for
A Bozkurt, B Esmaeili, DH Brooks, JG Dy, JW van de Meent. Can VAEs Generate Novel Examples?
In Critiquing and Correcting Trends in Machine Learning Workshop at NeurIPS 2018
The paper, poster, and slides are available at https://alicanb.github.io/publication/cract-vae/
If you use this code in your research please cite the paper:
@article{bozkurt2018can,
title={Can VAEs Generate Novel Examples?},
author={Bozkurt, Alican and Esmaeili, Babak and Brooks, Dana H and Dy, Jennifer G and van de Meent, Jan-Willem},
journal={arXiv preprint arXiv:1812.09624},
year={2018}
}
Training all models should take around 10 hours. Instead download the pretrained models:
python download_models.py
Codes for generating figures:
python plot_agg_posterior.py --num-points 1000
python plot_capacity_vs_reconstruction.py
python plot_kde_bce.py
python plot_compare_recons_wavg.py --enc-arch 400 200 100 --holdout
python plot_compare_recons_wavg.py --enc-arch 400 200 100 --holdout
