All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
The concerns raised by the reviewers have been addressed, so the paper can be accepted for publication in this journal. Congratulations to the authors!
N/A
N/A
N/A
The author has addressed most of my comments. I have no other concerns.
No comment.
No comment.
No comment.
This paper investigates the impact of network topology on the performance of an ADMM-based learning of Support Vector Machine (SVM) , and mean-degree graphs, and some common modern network topologies. Further, the paper investigates to which degree the expansion property of the network influences convergence for iterations, training and communication time. The experimental part is very sufficient. I have no doubts now.
This paper was reviewed by two experts. Both of them raised significant concerns about how ADMM is integrated with SVM. The authors are thus suggested to put efforts to explain the transformation/integration process.
[# PeerJ Staff Note: Please ensure that all review comments are addressed in a rebuttal letter and any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate. It is a common mistake to address reviewer questions in the rebuttal letter but not in the revised manuscript. If a reviewer raised a question then your readers will probably have the same question so you should ensure that the manuscript can stand alone without the rebuttal letter. Directions on how to prepare a rebuttal letter can be found at: https://peerj.com/benefits/academic-rebuttal-letters/ #]
.
.
.
This paper investigated the key features of networks for distributed ADMM in the applications of SVMs. Extensive experiments have been conducted for this purpose. However, there are still several problems need to be addressed.
1. As the ADMM is the key method used in this paper, it should appear in the title.
2. The section ADMM is confusing. The problems (12) and (13) are not formal. The process of the used consensus-based ADMM algorithm should be provided. It will be better if the convergence analysis could be given rather than only presenting the experiment stduies.
3. The transformation from SVM to ADMM is absent.
The wording of this article is more accurate, in line with the requirements of politeness and norms.This paper introduces the research background and current situation of the problem.The graphs are also clear.
This article should further explain how the Alternating Direction Method Of Multipliers (ADMM) algorithm can be applied to Support Vector Machine (SVM) training.Instead of just describing the Alternating Direction Method Of Multipliers algorithm and Support Vector Machine training model respectively.The article needs to add more details about the transformation between models.
(1)The simulation results in this paper are more comprehensive, and the conclusion of this paper is validated effectively.
(2)The computational part is persuasive.The parallel computations in this article are performed on Hebbe at Chalmers Centre for Computational 485 Science and Engineering (C3SE) provided by the Swedish National Infrastructure for Computing (SNIC).
(1)The paper should organize the references more strictly and unify the quotation format of multiple references.
(2)The language needs to pay more attention to details, such as the use of parentheses when referring to graphs.
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.