Dynamic Byzantine-Robust Learning: Adapting to Switching Byzantine Workers

Authors: Ron Dorfman, Naseem Amin Yehya, Kfir Yehuda Levy

ICML 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Finally, in Section 6, we explore the practical aspects and benefits of our approach through experiments on image classification tasks with two dynamic identity-switching strategies.
Researcher Affiliation Academia 1Department of Electrical and Computer Engineering, Technion, Haifa, Israel.
Pseudocode Yes Algorithm 1 Byzantine-Robust Optimization with MLMC
Open Source Code No The paper does not contain any statement or link providing concrete access to source code for the methodology described.
Open Datasets Yes We study image classification on the MNIST (Le Cun et al., 1998) and CIFAR-10 (Krizhevsky et al., 2009) datasets
Dataset Splits No The paper mentions using MNIST and CIFAR-10 datasets and reports test accuracy, but it does not explicitly provide the training, validation, and test dataset splits or their sizes.
Hardware Specification Yes We run all experiments on a machine with a single NVIDIA Ge Force RTX 4090 GPU.
Software Dependencies No The paper mentions using CNN architectures and training details but does not provide specific version numbers for any software components or libraries like Python, PyTorch, or CUDA.
Experiment Setup Yes Additional training details are deferred to Appendix J for brevity. ... Table 2. Training details and hyperparameters. ... Learning rate 10 drop after 4000 iterations 10 drop after 6000 iterations Weight decay 10 4 Base batch size 32 64