Improving Adversarial Robustness Through the Contrastive-Guided Diffusion Process

Authors: Yidong Ouyang, Liyan Xie, Guang Cheng

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

Reproducibility Variable Result LLM Response
Research Type Experimental We validate our theoretical results using simulations and demonstrate the good performance of Contrastive-DP on image datasets.
Researcher Affiliation Academia 1School of Data Science, The Chinese University of Hong Kong, Shenzhen, China 2Department of Statistics, University of California, Los Angeles, USA.
Pseudocode Yes Algorithm 1 Generation in Contrastive-guided Diffusion Process (Contrastive-DP)
Open Source Code No The paper does not explicitly provide a link to open-source code or state that the code will be made publicly available for the described methodology.
Open Datasets Yes We test the contrastive-DP algorithm on three image datasets, the MNIST dataset (Le Cun et al., 1998), CIFAR10 dataset (Krizhevsky, 2009), and the Traffic Signs dataset (Houben et al., 2013).
Dataset Splits No The paper specifies training and testing sizes for datasets like MNIST ("60k...for training and 10k...for testing"), but it does not explicitly mention validation dataset splits.
Hardware Specification Yes Running on a RTX 4x2080Ti GPU cluster.
Software Dependencies No The paper mentions tools and frameworks (e.g., PyTorch, TRADES, WRN-28-10) but does not provide specific version numbers for any software components.
Experiment Setup Yes A detailed description of the pipeline for generating data and the corresponding hyperparameter can be found in Appendix D.3.