Generating Adversarial Examples with Adversarial Networks

Authors: Chaowei Xiao, Bo Li, Jun-yan Zhu, Warren He, Mingyan Liu, Dawn Song

IJCAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In this section, we first evaluate Adv GAN for both semiwhitebox and black-box settings on MNIST [Le Cun and Cortes, 1998] and CIFAR-10 [Krizhevsky and Hinton, 2009]. We also perform a semi-whitebox attack on the Image Net dataset [Deng et al., 2009]. We then apply Adv GAN to generate adversarial examples on different target models and test the attack success rate for them under the state-of-the-art defenses and show that our method can achieve higher attack success rates compared to other existing attack strategies. We generate all adversarial examples for different attack methods under an L bound of 0.3 on MNIST and 8 on CIFAR-10, for a fair comparison. In general, as shown in Table 1, Adv GAN has several advantages over other white-box and blackbox attacks.
Researcher Affiliation Academia 1University of Michigan, Ann Arbor 2University of California, Berkeley 3Massachusetts Institute of Technology
Pseudocode No The paper describes processes such as dynamic distillation in a step-by-step manner, but it does not include any clearly labeled 'Pseudocode' or 'Algorithm' blocks.
Open Source Code No The paper provides links to third-party resources and challenges (e.g., 'https://github.com/Madry Lab/mnist_challenge', 'github.com/tensorflow/models/blob/master/research/Res Net'), but it does not include an explicit statement or link for the open-source code of their own proposed methodology (Adv GAN).
Open Datasets Yes In this section, we first evaluate Adv GAN for both semiwhitebox and black-box settings on MNIST [Le Cun and Cortes, 1998] and CIFAR-10 [Krizhevsky and Hinton, 2009]. We also perform a semi-whitebox attack on the Image Net dataset [Deng et al., 2009].
Dataset Splits No The paper mentions using well-known datasets like MNIST and CIFAR-10 which have standard train/test splits. However, it does not explicitly provide specific percentages or details for training, validation, and test splits used in their experiments, nor does it refer to a predefined validation split from a citation or external source.
Hardware Specification No The paper does not include any specific details about the hardware used for running the experiments (e.g., GPU models, CPU types, memory, or cloud instance specifications).
Software Dependencies No The paper mentions 'Tensor Flow' and 'Adam as our solver', but it does not specify version numbers for these or any other software dependencies needed to replicate the experiment.
Experiment Setup Yes We use Adam as our solver [Kinga and Adam, 2015], with a batch size of 128 and a learning rate of 0.001. We set the confidence κ = 0 for both Opt. and Adv GAN. We generate all adversarial examples for different attack methods under an L bound of 0.3 on MNIST and 8 on CIFAR-10, for a fair comparison.