Label Noise in Adversarial Training: A Novel Perspective to Study Robust Overfitting

Authors: Chengyu Dong, Liyuan Liu, Jingbo Shang

NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on different datasets, training methods, neural architectures and robustness evaluation metrics verify the effectiveness of our method.
Researcher Affiliation Collaboration Chengyu Dong University of California, San Diego cdong@eng.ucsd.edu Liyuan Liu Microsoft Research lucliu@microsoft.com Jingbo Shang University of California, San Diego jshang@eng.ucsd.edu
Pseudocode No The paper describes its methods in text and mathematical formulations but does not include any explicit pseudocode or algorithm blocks.
Open Source Code No The paper does not include an explicit statement about releasing source code or a link to a code repository for the described methodology.
Open Datasets Yes We conduct experiments on three datasets including CIFAR-10, CIFAR100 (Krizhevsky, 2009) and Tiny-Image Net (Le & Yang, 2015).
Dataset Splits No The paper mentions using a 'validation set' and 'training subset of size 5k' but does not provide specific split percentages or sample counts for the train/validation/test sets across all experiments to reproduce the data partitioning.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running the experiments.
Software Dependencies No The paper mentions various training methods and models (e.g., PGD training, Res Net-18, Auto Attack) but does not provide specific version numbers for any software dependencies or libraries used.
Experiment Setup Yes We conduct PGD training on pre-activation Res Net-18 (He et al., 2016) with 10 iterations and perturbation radius 8/255 by default.