Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..

Second-Order Convergence in Private Stochastic Non-Convex Optimization

Authors: Youming Tao, Zuyuan Zhang, Dongxiao Yu, Xiuzhen Cheng, Falko Dressler, Di Wang

NeurIPS 2025 | Venue PDF | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Numerical experiments on real-world datasets validate the efficacy of our approach.
Researcher Affiliation Academia Youming Tao TU Berlin & Shandong University EMAIL Zuyuan Zhang The George Washington University EMAIL Dongxiao Yu Shandong University EMAIL Xiuzhen Cheng Shandong University EMAIL Falko Dressler TU Berlin EMAIL Di Wang KAUST EMAIL
Pseudocode Yes Algorithm 1: Gauss-PSGD: Gaussian Perturbed Stochastic Gradient Descent Algorithm 2: Ada-DP-SPIDER Algorithm 3: Distributed Ada-DP-SPIDER Algorithm 4: Private Model Selection in Distributed Learning
Open Source Code No While our code is not yet released at the time of submission, we plan to open-source it with detailed instructions to reproduce all experimental results as described in the Appendix.
Open Datasets Yes The datasets used in our experiments are publicly available.
Dataset Splits No The paper states, "We provide a comprehensive description of our experimental setup, including running environments, datasets, learning models, hyperparameter settings, and evaluation metrics, in Section F of the Appendix." However, Section F is not provided in the given text, and no specific dataset split information is available in the main body.
Hardware Specification No The paper states, "We specify all the computational resources used for our experiments in the experiment section (Section F in Appendix)." However, Section F is not provided in the given text, and no specific hardware details are available in the main body.
Software Dependencies No The paper mentions that the experimental setup is detailed in Section F of the Appendix. However, Section F is not provided, and no specific software dependencies with version numbers are mentioned in the accessible text.
Experiment Setup No The paper states, "We provide a comprehensive description of our experimental setup, including running environments, datasets, learning models, hyperparameter settings, and evaluation metrics, in Section F of the Appendix." However, Section F is not provided in the given text, and no specific experimental setup details (like hyperparameters) are available in the main body.