Improving Adversarial Robustness by Putting More Regularizations on Less Robust Samples

Authors: Dongyoon Yang, Insung Kong, Yongdai Kim

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

Reproducibility Variable Result LLM Response
Research Type Experimental Numerical experiments illustrate that our proposed algorithm improves the generalization (accuracy on examples) and robustness (accuracy on adversarial attacks) simultaneously to achieve the state-of-the-art performance.
Researcher Affiliation Academia 1Department of Statistics, Seoul National University, Seoul, Republic of Korea.
Pseudocode Yes Algorithm 1 Anti-Robust Weighted (ARo W) Regularization
Open Source Code Yes The code is available at https://github.com/dyoony/ARoW.
Open Datasets Yes In this section, we investigate ARo W algorithm in view of robustness and generalization by analyzing the four benchmark data sets CIFAR10, CIFAR100 (Krizhevsky, 2009) , F-MINST (Xiao et al., 2017) and SVHN dataset (Netzer et al., 2011).
Dataset Splits No The paper mentions 'training dataset' and 'test data' but does not explicitly specify distinct training/validation/test splits with percentages or counts, or refer to predefined validation splits.
Hardware Specification No The paper does not provide specific details about the hardware (e.g., GPU models, CPU types, or memory) used for running the experiments.
Software Dependencies No The paper mentions the use of certain algorithms and optimizers (e.g., PGD, SGD) but does not specify any software libraries or dependencies with their version numbers (e.g., 'PyTorch 1.9', 'TensorFlow 2.x').
Experiment Setup Yes For generating adversarial examples in the training phase, PGD10 with random initial, p = , ε = 8/255 and ν = 2/255 is used... For training prediction models, the SGD with momentum 0.9, weight decay 5 × 10−4, the initial learning rate of 0.1 and batch size of 128 is used and the learning rate is reduced by a factor of 10 at 60 and 90 epochs. Stochastic weighting average (SWA) (Izmailov et al., 2018) is employed after 50-epochs... Table 11 presents the hyperparameters used on our experiments.