Collect at Once, Use Effectively: Making Non-interactive Locally Private Learning Possible

Authors: Kai Zheng, Wenlong Mou, Liwei Wang

ICML 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical In this paper, we propose several efficient algorithms for learning and estimation problems under non-interactive LDP model, with good theoretical guarantees.
Researcher Affiliation Academia 1Key Laboratory of Machine Perception, MOE, School of EECS, Peking University, Beijing, China. Correspondence to: Kai Zheng <zhengk92@pku.edu.cn>, Wenlong Mou <mouwenlong@pku.edu.cn>, Liwei Wang <wanglw@cis.pku.edu.cn>.
Pseudocode Yes Algorithm 1 Basic Private Vector mechanism ... Algorithm 2 LDP ℓ1 Constrained Mean Estimation ... Algorithm 3 LDP ℓ1 Constrained Linear Regression ... Algorithm 4 LDP kernel mechanism ... Algorithm 5 LDP SGLD Mechanism Collection ... Algorithm 6 LDP SGLD Mechanism Learning
Open Source Code No The paper does not include any statement or link indicating that the source code for the described methodology is publicly available.
Open Datasets No The paper is theoretical and does not conduct experiments on a specific publicly available dataset. It defines problem settings with data properties (e.g., 'distribution D supported on B(0, 1)') but does not provide access information for empirical data.
Dataset Splits No The paper is theoretical and does not present empirical experiments, therefore it does not provide training/test/validation dataset splits.
Hardware Specification No The paper is theoretical and does not describe any experimental setup or specific hardware used for running experiments.
Software Dependencies No The paper is theoretical and does not describe any specific software dependencies with version numbers for replication.
Experiment Setup No The paper is theoretical and does not include details about an experimental setup such as hyperparameters or training configurations.