Run-off Election: Improved Provable Defense against Data Poisoning Attacks

Authors: Keivan Rezaei, Kiarash Banihashem, Atoosa Chegini, Soheil Feizi

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

Reproducibility Variable Result LLM Response
Research Type Experimental We evaluate our methods on MNIST, CIFAR-10, and GTSRB and obtain improvements in certified accuracy by up to 3%-4%.
Researcher Affiliation Academia 1Department of Computer Science, University of Maryland, MD, USA.
Pseudocode Yes The formal pseudocode of ROE is provided in Algorithm 1.
Open Source Code Yes Our code can be found in this github repository.
Open Datasets Yes We similarly use Network-In-Network (Lin et al., 2013) architecture, to be trained with the set of hyperparameters from (Gidaris et al., 2018). We similarly use Network-In-Network (Lin et al., 2013) architecture, to be trained with the set of hyperparameters from (Gidaris et al., 2018).
Dataset Splits No The paper mentions 'training data' and 'test samples' and states 'We consider the same setup as prior work (Levine & Feizi, 2020; Wang et al., 2022b)', implying standard splits. However, it does not explicitly describe a separate 'validation' split or its specific percentages/counts.
Hardware Specification Yes by using a single NVIDIA Ge Force RTX 2080 Ti GPU
Software Dependencies No The paper does not specify version numbers for any software or libraries used (e.g., Python, PyTorch, CUDA).
Experiment Setup Yes We similarly use Network-In-Network (Lin et al., 2013) architecture, to be trained with the set of hyperparameters from (Gidaris et al., 2018).