Secure Federated Correlation Test and Entropy Estimation
Authors: Qi Pang, Lun Wang, Shuai Wang, Wenting Zheng, Dawn Song
ICML 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The evaluation results demonstrate that FED-χ2 and FED-H achieve good performance with small client-side computation overhead in several real-world case studies. |
| Researcher Affiliation | Collaboration | 1Carnegie Mellon University 2Google 3Hong Kong University of Science and Technology 4UC Berkeley. |
| Pseudocode | Yes | Algorithm 1 The encoding and decoding scheme (Indyk, 2006) for federated frequency moments estimation. |
| Open Source Code | Yes | We maintain our code at https://github.com/ Qi-Pang/Federated-Correlation-Test. |
| Open Datasets | Yes | We evaluate FED-χ2 on 4 synthetic datasets and 16 real-world datasets. The details for the real-world datasets used in Sec. 4.1 are provided in Table 1. The license of Credit Risk Classification (Govindaraj, Praveen) is CC BY-SA 4.0, the license of German Traffic Sign (Houben et al., 2013) is CC0: Public Domain. Other datasets without a license are from UCI Machine Learning Repository (Dua & Graff, 2017). URL http://archive.ics.uci.edu/ml. |
| Dataset Splits | No | The paper specifies training and test splits, but no explicit validation split is mentioned for reproducibility. |
| Hardware Specification | Yes | Unless otherwise specified, experiments are launched on an Ubuntu 18.04 LTS server equipped with 32 AMD Opteron(TM) Processor 6212 and 512GB RAM. ... All models are trained on NVIDIA Ge Force RTX 3090. |
| Software Dependencies | No | The paper mentions "Py Droid (Sandeep Nandal, 2020)" but does not provide specific version numbers for other key software components, libraries, or programming languages used for the experiments. |
| Experiment Setup | Yes | We set the p-value threshold as 0.05. ... The hyper-parameters for the SAFFRON procedure in online false discovery rate control of Sec. 4 are aligned with the setting in (Ramdas et al., 2018). The target FDR level is α = 0.05, the initial wealth is W0 = 0.0125, and γj is calculated in the following way: γj = 1/(j+1)1.6 P10000 j=0 1/(j+1)1.6 . ... All hyper-parameters are the same. The details of these models are reported in Appendix K. ... We use the same learning rate; random seed and all other settings are also the same to make the comparison fair. |