Identification of the Adversary from a Single Adversarial Example

Authors: Minhao Cheng, Rui Min, Haochen Sun, Pin-Yu Chen

ICML 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Finally, the effectiveness of our proposed framework is evaluated by extensive experiments with different model architectures, adversarial attacks, and datasets.
Researcher Affiliation Collaboration 1Department of Computer Science & Engineering, The Hong Kong University of Science and Technology, Hong Kong 2IBM Research, NY, USA.
Pseudocode No The paper describes methods in narrative text and does not include explicit pseudocode or algorithm blocks.
Open Source Code Yes Our code is publicly available at https://github.com/ rmin2000/adv_tracing.git.
Open Datasets Yes We conduct our experiments on three popular image classification datasets GTSRB (Stallkamp et al., 2012), CIFAR-10 (Krizhevsky et al., 2009) and Tiny-Image Net (Le & Yang, 2015).
Dataset Splits No The paper mentions using CIFAR-10, GTSRB, and Tiny-Image Net datasets but does not explicitly state the train/validation/test dataset splits with percentages, absolute sample counts, or explicit references to predefined splits with citations.
Hardware Specification Yes All our experiments were implemented in Pytorch and conducted using an RTX 3090 GPU.
Software Dependencies No The paper mentions software like Pytorch and Adversarial Robustness Toolbox (ART) but does not provide specific version numbers for these dependencies.
Experiment Setup Yes In the pretraining stage, the base model is trained with Adam optimizer with the learning rate of 10 3 and a batch size of 128 for 50 epochs. For every copy s watermark, we independently randomly sample 100 pixels to mask for both CIFAR-10 and GTSRB and increase the mask size to 400 pixels for Tiny-Image Net which keep the masking rate around 3.26%.