chop
chop copied to clipboard
openopt
CHOP: An optimization library based on PyTorch, with applications to adversarial examples and structured neural network training.
See project proposal [here](https://andre-martins.github.io/pages/project-examples-for-deep-structured-learning-fall-2019.html).
Replace $\max_\delta \mathcal L(f_\theta(x + \delta), y)$ with $\max_{g\in \mathcal G} L(f_\theta(g(x), y))$ Example: filter based models where $g$ is a piece-wise linear function.
Related to curriculum training: make examples easier at first by minimizing over input space. Try to do so especially for hard examples.