Adapting to function difficulty and growth conditions in private optimization

Authors: Hilal Asi, Daniel Levy, John C. Duchi

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

Reproducibility Variable Result LLM Response
Research Type Theoretical We develop algorithms for private stochastic convex optimization that adapt to the hardness of the specific function we wish to optimize. ... Our algorithms build upon the inverse sensitivity mechanism... We complement our algorithms with matching lower bounds... (Self-assessment: If you ran experiments... [N/A])
Researcher Affiliation Academia Hilal Asi Daniel Levy John C. Duchi {asi,danilevy,jduchi}@stanford.edu
Pseudocode Yes Algorithm 1 Localization-based Algorithm and Algorithm 2 Epoch-based algorithms for κ-growth are present in the paper.
Open Source Code No The paper does not contain any explicit statements about releasing source code or links to a code repository. The self-assessment under 'If you are including theoretical results...' and 'If you ran experiments...' sections mark code availability as '[N/A]'.
Open Datasets No The paper is theoretical and does not describe experiments using datasets. The self-assessment questions for 'If you ran experiments...' and 'If you are using existing assets...' are all marked as '[N/A]' regarding data.
Dataset Splits No The paper is theoretical and does not describe experiments, thus it does not provide details on training, validation, or test dataset splits. The self-assessment questions for 'If you ran experiments...' are all marked as '[N/A]'.
Hardware Specification No The paper is theoretical and does not describe any experiments, therefore it does not specify any hardware used. The self-assessment question 'Did you include the total amount of compute and the type of resources used?' is marked '[N/A]'.
Software Dependencies No The paper is theoretical and does not describe any experiments, therefore it does not specify any software dependencies with version numbers. The self-assessment questions related to experiments are marked '[N/A]'.
Experiment Setup No The paper is theoretical and does not describe any experiments, therefore it does not provide specific experimental setup details such as hyperparameters or training configurations. The self-assessment questions related to experiments are marked '[N/A]'.