ARMOURED: Adversarially Robust MOdels using Unlabeled data by REgularizing Diversity

Authors: Kangkang Lu, Cuong Manh Nguyen, Xun Xu, Kiran Krishnamachari, Yu Jing Goh, Chuan-Sheng Foo

ICLR 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental 4 EXPERIMENTS
Researcher Affiliation Academia 1 Institute for Infocomm Research, A*STAR, Singapore 2 National University of Singapore, Singapore
Pseudocode Yes Pseudocode detailing the training procedure is provided in Algorithm 1 in Appendix A.1.
Open Source Code No The paper does not contain an explicit statement about releasing source code or a link to a code repository.
Open Datasets Yes We evaluate ARMOURED on the CIFAR-10 and SVHN datasets. We use the official train/test splits (50k/10k labeled samples) for CIFAR-10 (Krizhevsky et al., 2009) and reserve 5k samples from the training samples for a validation set. In our semi-supervised setup, the label budget is either 1k or 4k; remaining samples from training set are treated as unlabeled samples. For the SVHN dataset (Netzer et al., 2011), our train/validation/test split is 65,932 / 7,325 / 26,032 samples.
Dataset Splits Yes We use the official train/test splits (50k/10k labeled samples) for CIFAR-10 (Krizhevsky et al., 2009) and reserve 5k samples from the training samples for a validation set. In our semi-supervised setup, the label budget is either 1k or 4k; remaining samples from training set are treated as unlabeled samples. For the SVHN dataset (Netzer et al., 2011), our train/validation/test split is 65,932 / 7,325 / 26,032 samples.
Hardware Specification No The paper does not specify the exact hardware (e.g., GPU/CPU models, memory) used for running its experiments, only mentioning the use of a Wide ResNet backbone.
Software Dependencies No The paper mentions software components like 'batch normalization', 'leaky ReLU activation', and 'Adam optimizer', but it does not provide specific version numbers for any software dependencies.
Experiment Setup Yes We train each method for 600 epochs on CIFAR-10-semi-4k and SVHN-semi-1k. Learning rate is decayed by a factor of 0.2 after the first 400k iterations. For the AT wrapper, we apply a 7-step PGD ℓ attack with total ϵ = 8/255 (for CIFAR-10), ϵ = 4/255 (for SVHN) and step size of ϵ/4. After tuning, we decide to apply (λDPP, λNEM) = (1, 1) for SVHN and (λDPP, λNEM) = (1, 0.5) for CIFAR-10.