The Price of Differential Privacy under Continual Observation

Authors: Palak Jain, Sofya Raskhodnikova, Satchit Sivakumar, Adam Smith

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

Reproducibility Variable Result LLM Response
Research Type Theoretical We provide the first strong lower bounds on the error of continual release mechanisms. In particular, for two fundamental problems that are closely related to empirical risk minimization and widely studied and used in the standard (batch) model, we prove that the worst case error of every continual release algorithm is Ω(T 1/3) times larger than that of the best batch algorithm.
Researcher Affiliation Academia 1Department of Computer Science, Boston University, Boston, Massachusetts, USA.
Pseudocode Yes Algorithm 1 Privacy game ΠM,Adv for the continual release model with adaptively chosen inputs
Open Source Code No The paper does not provide any statement or link indicating the release of open-source code for the described methodology.
Open Datasets No The paper is theoretical and focuses on proofs and algorithm design, not empirical evaluation on datasets. Therefore, it does not mention publicly available datasets for training.
Dataset Splits No The paper is theoretical and does not conduct empirical experiments with datasets, so it does not specify training, validation, or test splits.
Hardware Specification No The paper is theoretical and does not involve experimental evaluation requiring specific hardware. Therefore, no hardware specifications are mentioned.
Software Dependencies No The paper is theoretical and does not describe empirical experiments with specific software implementations or dependencies.
Experiment Setup No The paper is theoretical and does not describe empirical experiments, thus no experimental setup details like hyperparameters or training settings are provided.