Communication Efficient Distributed Newton Method over Unreliable Networks

Authors: Ming Wen, Chengchang Liu, Yuedong Xu

AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results demonstrate its advantage over first-order and second-order baselines, and its tolerance to packet loss rate ranging from 5% to 40%.
Researcher Affiliation Academia Ming Wen1, Chengchang Liu2, Yuedong Xu1 1Fudan University, 2The Chinese University of Hong Kong mwen23@m.fudan.edu.cn, ccliu22@cse.cuhk.hk, ydxu@fudan.edu.cn
Pseudocode Yes Algorithm 1: RED-New
Open Source Code No The paper does not provide any statement or link indicating that the source code for the described methodology is publicly available.
Open Datasets Yes Our experiments are conducted on the benchmark LIBSVM Libraray datasets (Chang and Lin 2011)
Dataset Splits No The paper mentions using LIBSVM datasets but does not explicitly provide specific training/validation/test dataset split percentages, counts, or cite a predefined split used for reproduction.
Hardware Specification Yes The experiments are operated on the Intel(R) Xeon(R) Platinum 8369B 2.90GHz, equipped with 32 CPU cores and 2.0TB memory.
Software Dependencies Yes We conduct our experiments using Python with the Mpy4pi 3.1.4 function for distributed optimization.
Experiment Setup Yes The learning rate of the proposed methods and the baselines are tuned from {1, 0.1, 0.01, 0.001}.