IBM / Autozoom-Attack

Codes for reproducing query-efficient black-box attacks in “AutoZOOM: Autoencoder-based Zeroth Order Optimization Method for Attacking Black-box Neural Networks” ​​​​, published at AAAI 2019
https://arxiv.org/abs/1805.11770
Apache License 2.0
57 stars 22 forks source link

What shoult I do if I want to use this algorithm in orther models and dataset? #16

Open chl2 opened 4 years ago

chl2 commented 4 years ago

I am re-implementing your code in other dataset like Imagenet64, other models like resnet50. How you train the encoder_weight_file and decoder_weight_file ? Can I use it directly? Thank You.

chunchentu commented 4 years ago

Hi,

Thanks for being interested in our work. The information of the codec is defined in setup_codec. You can then train encoder/decoder based on the model definition.