RECESS Vaccine for Federated Learning: Proactive Defense Against Model Poisoning Attacks

Authors: Haonan Yan, Wenjing Zhang, Qian Chen, Xiaoguang Li, Wenhai Sun, HUI LI, Xiaodong Lin

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

Reproducibility Variable Result LLM Response
Research Type Experimental Finally, we extensively evaluate RECESS on typical model architectures and four datasets under various settings... Experimental results show the superiority of RECESS in terms of reducing accuracy loss caused by the latest model poisoning attacks over five classic and two state-of-the-art defenses.
Researcher Affiliation Academia 1Xidian University, 2University of Guelph, 3Purdue University
Pseudocode No The paper includes a section titled 'Proactive Detection Algorithm' describing steps, but it is not formatted as a pseudocode block or a clearly labeled algorithm.
Open Source Code No The paper does not contain an explicit statement about releasing its source code, nor does it provide a direct link to a code repository for the methodology described.
Open Datasets Yes Table 1 shows four datasets and parameter settings used in the evaluation. ... MNIST, CIFAR-10, Purchase, FEMNIST.
Dataset Splits No The paper mentions using IID and Non-IID dataset divisions and references other works for dataset construction methods, but it does not explicitly provide specific percentages or counts for training, validation, or test dataset splits.
Hardware Specification No The paper does not provide any specific details about the hardware (e.g., GPU models, CPU types, memory) used to conduct the experiments.
Software Dependencies No The paper mentions optimizers (Adam, SGD) but does not provide specific software names with version numbers for libraries, frameworks, or environments used in the experiments.
Experiment Setup Yes Table 1: Experiment datasets and FL settings. ... Dataset... Model... Clients... Batch Size... Optimizer... Learning Rates... Epochs. For RECESS, we set A = 0.95, TS0 = 1, and baseline_decreased_score = 0.1 unless otherwise specified.