making adversarial examples against ML/DL methods used in astrophysics
- David W Hogg (NYU) (MPIA) (Flatiron)
- Soledad Villar (NYU)
All content copyright 2019 the authors.
The code herein is released open-source under the MIT License.
See the file LICENSE
for more details.
- Comparing generative models to standard machine-learning methods in terms of adversarial robustness.
- Generalizing adversarial approaches to regression settings.
- Relationships between the success of adversaries and the concepts of overfitting, generalizability, and interpretability.