Zhakshylyk Nurlanov

Hello, Is it possible to compute bounds for a batch of inputs with different epsilon bounds? I discovered it could be done for L_inf norm, but non-trivial for other norms....

@huanzhang12 However, such obtained bounds will not be the same as in CROWN method, won't they? I wanted to look at intermediate bounds from particular methods. Is it possible to...

Other distributions for IBP init

1

Hi, Have you tried other distributions for IBP init? Does it make difference? Since the IBP training has non-continuity at `W_i = 0`, would it make sense to initialize it...

Multiple parallel runs on multi-GPU for hyperparameter search

4

Hi, I have found that the `--multirun` option with multiple parameters does not work out of box on [LSF](https://www.ibm.com/docs/en/spectrum-lsf/10.1.0?topic=jobs-how-lsf-runs-parallel) cluster. E.g. requesting 3 gpus and running: ``` #BSUB -gpu "num=3"...

Resume multi-run

3

Hi, Is it possible to resume a multi-run? E.g. if the Optuna hyperparameter search has crashed, can we resume the search from that point without having to sample new runs?

Hi, Could you also provide hyperparameters for training IBP models (MNIST, CIFAR) to get the same numbers as in the paper? Or pretrained models? I would like to reproduce those...

Evaluating MNTP task only

1

Thanks for your work. I am wondering if you plan to provide an example script for loading and evaluating the MNTP task only. In my experiments, the results are `_attn_implementation`...

Including standard accuracies

1

Hi, Thanks for maintaining this great repo and the leaderboard. It would be great to also include the standard accuracies in the leaderboard (similar to https://robustbench.github.io/ ) to fairly assess...