Loss weights for different datasets

aamir-mustafa / pcl-adversarial-defense

Adversarial Defense by Restricting the Hidden Space of Deep Neural Networks, in ICCV 2019

59 stars 14 forks source link

Closed krishnakanthnakka closed 4 years ago

krishnakanthnakka commented 4 years ago

Can you please share the weight-prox,weight-conprox, lr_prox, lr_conprox, lr_model for different datasets.

I find that the experiments are extremely sensitive to prox/conprox related parameters. This will be helpful for reproducing results.

aamir-mustafa commented 4 years ago

Hi,

Yes I am planning to share a separate file containing the hyper-parameters chosen while training for all 5 datasets used in the paper.

Thanks for the interest