Project | Arxiv

Accepted to NeurIPS'22: "Semantic uncertainty intervals for disentangled latent spaces"

We obtain conformalized uncertainty intervals for generative models with semantic meaning!

Coming soon:

• Add pretrained models for CELEB-A, CLEVR, LSUN Cars
• Add gradio demo
• Add colab notebook demo
• Add training code

This repository contains evaluation code built on top of pixel2style2pixel. Please follow the instructions to install pixel2style2pixel for setting up this repository.

The project is licensed under MIT License.

The research for this project was sponsored by the U.S. Air Force Research Laboratory and the U.S. Air Force Artificial Intelligence Accelerator and was accomplished under Cooperative Agreement Number FA8750-19-2- 1000. MIT SuperCloud and the Lincoln Laboratory Supercomputing Center also provided computing resources that contributed to the results reported in this work.