Distributed Saddle-Point Problems Under Data Similarity

Authors: Aleksandr Beznosikov, Gesualdo Scutari, Alexander Rogozin, Alexander Gasnikov

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

Reproducibility Variable Result LLM Response
Research Type Experimental We simulate the Robust Linear Regression problem... We assess the effectiveness of the proposed algorithms on a robust regression problem. Figure 1 compares the performance of Algorithm 1 and the Centralized Extragradient method... Our second experiment is using real data, specifically LIBSVM datasets [8].
Researcher Affiliation Collaboration Aleksandr Beznosikov MIPT , HSE University and Yandex, Russia Gesualdo Scutari Purdue University, USA Alexander Rogozin MIPT and HSE University, Russia Alexander Gasnikov MIPT, HSE University and ISP RAS , Russia
Pseudocode Yes Algorithm 1 (Star Min-Max Data Similarity Algorithm) ... Algorithm 2 (Distributed Min-Max Data Similarity Algorithm) ... Algorithm 3 (Acc Gossip)
Open Source Code Yes Source code: https://github.com/alexrogozin12/data_sim_sp
Open Datasets Yes Our second experiment is using real data, specifically LIBSVM datasets [8].
Dataset Splits No No specific training/validation/test dataset splits (percentages, counts, or explicit standard split citations) were found in the paper's main text for reproducibility.
Hardware Specification No No specific hardware details (like GPU/CPU models, processor types, or memory amounts) used for running the experiments were found in the paper.
Software Dependencies Yes The algorithms are implemented in Python 3.73.
Experiment Setup Yes A description of the tuning of the algorithm parameters can be found in Appendix C.