Provably Secure Federated Learning against Malicious Clients

Authors: Xiaoyu Cao, Jinyuan Jia, Neil Zhenqiang Gong6885-6893

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

Reproducibility Variable Result LLM Response
Research Type Experimental We evaluate our method on MNIST and Human Activity Recognition datasets. For instance, our method can achieve a certified accuracy of 88% on MNIST when 20 out of 1,000 clients are malicious. Evaluation: We evaluate our methods on MNIST and Human Activity Recognition datasets.
Researcher Affiliation Academia Xiaoyu Cao, Jinyuan Jia, Neil Zhenqiang Gong Duke University, Durham, NC 27708 {xiaoyu.cao, jinyuan.jia, neil.gong}@duke.edu
Pseudocode Yes Algorithm 1 Single-global-model federated learning; Algorithm 2 Computing Predicted Label and Certified Security Level
Open Source Code No The paper does not contain any explicit statements about releasing source code for the methodology or provide a link to a code repository.
Open Datasets Yes We use MNIST (Le Cun, Cortes, and Burges 1998) and Human Activity Recognition (HAR) datasets (Anguita et al. 2013).
Dataset Splits No The paper mentions 60,000 training examples and 10,000 testing examples for MNIST, and for HAR, '75% of each user s examples as training examples and the rest as testing examples.' There is no explicit mention of a separate validation split percentage or count.
Hardware Specification No The paper does not specify the hardware (e.g., GPU/CPU models, memory, or cloud instances) used for running the experiments.
Software Dependencies No The paper mentions using Fed Avg and provides hyperparameters but does not list specific software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow versions).
Experiment Setup Yes Table 1: Federated learning settings and hyperparameters. provides specific values for 'global Iter', 'local Iter', 'Learning rate η', and 'Batch size'. The text also states: 'In particular, we set the global Iter in Table 1 because Fed Avg converges with such settings.'