Next Article in Journal
Nonlinear Dynamic Response of a CC-RCC Combined Dam Structure under Oblique Incidence of Near-Fault Ground Motions
Previous Article in Journal
Lightweight Attention Pyramid Network for Object Detection and Instance Segmentation
 
 
Article
Peer-Review Record

DE-CapsNet: A Diverse Enhanced Capsule Network with Disperse Dynamic Routing

Appl. Sci. 2020, 10(3), 884; https://doi.org/10.3390/app10030884
by Bohan Jia and Qiyu Huang *
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Appl. Sci. 2020, 10(3), 884; https://doi.org/10.3390/app10030884
Submission received: 24 December 2019 / Revised: 20 January 2020 / Accepted: 22 January 2020 / Published: 29 January 2020
(This article belongs to the Section Computing and Artificial Intelligence)

Round 1

Reviewer 1 Report

This paper presents good results interesting for the journal readers in the field of network and computational science, however there are few issues that should be corrected before acceptance:

Part 2, related works is still poor. For supporting this work other descriptors and references are needed. I saw many references related to this topic which are not discussed here. In Fig.1 the objects and letters are too small. They should be increased in proportion to the text. Some of Eq. 1 to Eq. 8 need some source reference and they must be cited in order to reinforce the work. 2 and Fig.3 are low quality, should be improved. In Fig.4 it is impossible to read the letters, they are too small. It is not clear in this paper which are the specific applications of the work, and which are the future perspectives in order to reduce the computational complexity and make it more reliable.

Comments for author File: Comments.pdf

Author Response

Dear Reviewer 1,

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper seems to be good, but you present your proposals unclear and expeditious. Algorithms are framed in tables, which denotes that the paper was written a little bit in haste. I propose you to rewrite the work more carefully. I may be wrong, but you are not convincing of the value of your model.

Author Response

Dear reviewer2,

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The paper is better written. I remain with the opinion that an algorithm cannot be presented inside a table. But this is a minor thing. The equations are numbered incorrectly. Equation 8 is missing.

The references are questionable. I counted 21 references from the arXiv.

The function of calculating the routing coefficients is interesting. The strengths of the paper are the proposed architecture and the reduction of the number of parameters to be calculated. 

I appreciate the knowledge of the subject and the practical skills for experiments.

Author Response

Dear Reviewer,

Please see the attachment

Author Response File: Author Response.pdf

Back to TopTop