search

fra31 / auto-attack

Code relative to "Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attacks"
https://arxiv.org/abs/2003.01690
MIT License
642 stars 111 forks source link

Add LBGAT on CIFAR-10 #36

Closed jiequancui closed 3 years ago

jiequancui commented 3 years ago

Paper: {Learnable Boundary Guided Adversarial Training; https://arxiv.org/pdf/2011.11164.pdf}

Venue: {if applicable, the venue where the paper appeared}

Dataset and threat model: {CIFAR-10, L-inf and epsilon 0.031}

Code: { code}

Pre-trained model: LBGAT0-wideresnet-34-10, LBGAT0-wideresnet-34-20

Log file: {LBGAT0-wideresnet-34-10, LBGAT0-wideresnet-34-20}

Additional data: {no}

Clean and robust accuracy: {88.22/52.86, 88.70/53.57}

Architecture: {wideresnet-34-10; wideresnet-34-20}

Description of the model/defense: {Our method aims to enhance model robustness while preserving high natural accuracy. }

fra31 commented 3 years ago

Hi,

thanks for submission! I've added the models.

jiequancui commented 3 years ago

Very Thanks.

jiequancui commented 3 years ago

Hi,

Our paper "Learnable boundary guided adversarial training" is to appear in ICCV2021. Can you help update the "Venue status" for our LBGAT models in Robust Bench?