Compressing deep graph convolution network with multi-staged knowledge distillation

Junghun Kim; Jinhong Jung; U. Kang

doi:10.1371/journal.pone.0256187

Peer Review History

Original SubmissionMay 13, 2021
28 May 2021 Decision Letter - Thippa Reddy Gadekallu, Editor PONE-D-21-15818 MustaD: Compressing Deep Graph Convolution Network with Multi-Staged Knowledge Distillation PLOS ONE Dear Dr. Kang, Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process. ============================== ACADEMIC EDITOR: Based on the comments received from the reviewers and my own observation, I recommend major revisions for the paper. The authors should carefully address all the comments and suggestions from the reviewers. Also, the authors should proofread to polish the English grammar in the paper. ============================== Please submit your revised manuscript by Jul 12 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file. Please include the following items when submitting your revised manuscript: A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'. A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'. An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'. If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter. If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols. We look forward to receiving your revised manuscript. Kind regards, Thippa Reddy Gadekallu Academic Editor PLOS ONE Journal Requirements: When submitting your revision, we need you to address these additional requirements. 1) Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf 2) Please amend either the abstract on the online submission form (via Edit Submission) or the abstract in the manuscript so that they are identical. [Note: HTML markup is below. Please do not edit.] Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented. Reviewer #1: Yes Reviewer #2: Yes ******** 2. Has the statistical analysis been performed appropriately and rigorously? Reviewer #1: Yes Reviewer #2: Yes ****** 3. Have the authors made all data underlying the findings in their manuscript fully available? The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified. Reviewer #1: Yes Reviewer #2: Yes ****** 4. Is the manuscript presented in an intelligible fashion and written in standard English? PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here. Reviewer #1: Yes Reviewer #2: Yes ****** 5. Review Comments to the Author Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters) Reviewer #1: 1. Does the proposed model has the ability to gain the full knowledge about all the features when there are multiple layers? 2. Does the author considers the semantic problems of features in the proposed method? 3. What happens to the Knowledge Distillation when the teacher’s accuracy is too low? Does the proposed system helps in solving these issues? 4. In the process of KD, Does the training errors in teacher propagate to student directly? Can the proposed system identify the errors? Please cite the following papers 1. Ashokkumar P, Siva Shankar G, Gautam Srivastava, Praveen Kumar Reddy Maddikunta, and Thippa Reddy Gadekallu. 2021. A Two-stage Text Feature Selection Algorithm for Improving Text Classification. ACM Trans. Asian Low-Resour. Lang. Inf. Process. 20, 3, Article 49 (April 2021), 19 pages. DOI:https://doi.org/10.1145/3425781 2. G. Siva Shankar, P. Ashokkumar, R. Vinayakumar, Uttam Ghosh, Wathiq Mansoor, Waleed S. Alnumay, "An Embedded-Based Weighted Feature Selection Algorithm for Classifying Web Document", Wireless Communications and Mobile Computing, vol. 2020, Article ID 8879054, 10 pages, 2020. https://doi.org/10.1155/2020/8879054 Reviewer #2: - The quality of the figures can be improved more. Figures should be eye-catching. It will enhance the interest of the reader. The abstract is long and NOT satisfactory. It should contain the following parts: i. The importance of or motivation for the research. ii. The issue/argument of the research. iii. The methodology. iv. The result/findings. v. The implications of the result/findings. - Please highlight the contribution clearly in the introduction - In the first four paragraphs of literature review section, the authors have presented a good references, but they need to present the recent and most updated references. - In the literature review section, you need to be consistent in the use of the verb tense, it is common to use the past tenses. - The summary at the end of the literature review should be focused on the limitations of related work. - The discussion is very important in research paper. Nevertheless, this section is short and should be presented completely. - Major contribution was not clearly mentioned in the conclusion part. - Make sure the Conclusion succinctly summarizes the paper. It should not repeat phrases from the Introduction! - Authors should add the most recent reference: 1) CANintelliIDS: Detecting In-Vehicle Intrusion Attacks on a Controller Area Network using CNN and Attention-based GRUCANintelliIDS: Detecting In-Vehicle Intrusion Attacks on a Controller Area Network using CNN and Attention-based GRU 2) DeepAMD: Detection and identification of Android malware using high-efficient Deep Artificial Neural Network, Future Generation Computer Systems 115, 844-856 ****** 6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review?** For information about this choice, including consent withdrawal, please see our Privacy Policy. Reviewer #1: No Reviewer #2: No [NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.] While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step. https://doi.org/10.1371/journal.pone.0256187.r001
Revision 1
6 Jul 2021 Author Response Reviewer #1 (R1-1) Does the proposed model has the ability to gain the full knowledge about all the features when there are multiple layers? – (A1-1) Accuracy of MustaD improves as the number of GCN layers in the teacher increases. This implies that MustaD gains more knowledge from the given input features when more GCN layers are used in the teacher. In other words, MustaD has the ability to gain the full knowledge about all the features when there are multiple layers in the teacher. We reported this in lines 334-336 in experiments section. • (R1-2) Does the author considers the semantic problems of features in the proposed method? – (A1-2) Since the main purpose of MustaD is to compress a deep GCN model to a compact model while preserving the multi-hop feature aggregation of the deep model, it is hard to expect that MustaD considers the semantic problems of features. We added the discussion on this as a future work in conclusion section (lines 447-448). • (R1-3) What happens to the Knowledge Distillation when the teachers accuracy is too low? Does the proposed system helps in solving these issues? – (A1-3) Through the experiments summarized in Table 3, we have shown that MustaD has an ability to aggregate information from farther nodes than the teacher, thus relieving the student’s strong dependence on the performance of the teacher. We reported this in lines 341-345 in experiments section. • (R1-4) In the process of KD, does the training errors in teacher propagate to student directly? Can the proposed system identify the errors? – (A1-4) Fig 4 shows that MustaD compensates for the training error of merely propagating the prediction of the teacher. We discussed this in lines 420-425 in experiments section. • (R1-5) Please cite the following papers: 1) Ashokkumar P, Siva Shankar G, Gautam Srivastava, Praveen Kumar Reddy Maddikunta, and Thippa Reddy Gadekallu. 2021. A Two-stage Text Feature Selection Algorithm for Improving Text Classification. ACM Trans. Asian Low-Resour. Lang. Inf. Process. 20, 3, Article 49 (April 2021), 19 pages. DOI:https://doi.org/10.1145/3425781, and 2) G. Siva Shankar, P. Ashokkumar, R. Vinayakumar, Uttam Ghosh, Wathiq Mansoor, Waleed S. Alnumay, ”An Embedded-Based Weighted Feature Selection Algorithm for Classify- ing Web Document”, Wireless Communications and Mobile Computing, vol. 2020, Article ID 8879054, 10 pages, 2020. https://doi.org/10.1155/2020/8879054 – (A1-5) We added the two references (lines 65-67 in related work section). Reviewer #2 (R2-1) The quality of the figures can be improved more. Figures should be eye-catching. It will enhance the interest of the reader. – (A2-1) We improved Figures 1 and 2. In particular, we redrew Fig 1 to be more informative, which previously looked complicated, and added markups for Fig 2 to be eye-catching. • (R2-2) The abstract is long and NOT satisfactory. It should contain the following parts: 1) The importance of or motivation for the research, 2) The issue/argument of the research, 3) The methodology, 4) The result/findings, and 5) The implications of the result/findings. – (A2-2) We rewrote the abstract to contain those five parts. • (R2-3) Please highlight the contribution clearly in the introduction. – (A2-3) We revised introduction section to emphasize the contribution of MustaD (lines 32-43 in introduction section). Furthermore, the contents of the contribution list have also been revised to clearly highlight the contributions (lines 54-56 in introduction section). • (R2-4) In the first four paragraphs of literature review section, the authors have presented a good references, but they need to present the recent and most updated references. – (A2-4) We added four up-to-date references (lines 65-67 in related work section). • (R2-5) In the literature review section, you need to be consistent in the use of the verb tense, it is common to use the past tenses. – (A2-5) We reviewed the literature review section to be consistent in the use of the verb tense. • (R2-6) The summary at the end of the literature review should be focused on the limitations of related work. – (A2-6) We added the limitations of deep GCNs at the end of the literature review (lines 104-106 in related work section). • (R2-7) The discussion is very important in research paper. Nevertheless, this section is short and should be presented completely. – (A2-7) We included more complete discussion (lines 334-336, lines 341-345, lines 380-387, lines 420-425, and lines 430-433 in experiments section). • (R2-8) Major contribution was not clearly mentioned in the conclusion part. – (A2-8) We revised the conclusion part to clearly present our major contribution. • (R2-9) Make sure the Conclusion succinctly summarizes the paper. It should not repeat phrases from the Introduction! – (A2-9) We revised conclusion section so that the conclusion succinctly summarizes the paper. • (R2-10) Authors should add the most recent reference: 1) CANintelliIDS: Detecting In-Vehicle Intrusion Attacks on a Controller Area Network using CNN and Attention-based GRU, and 2) DeepAMD: Detection and identification of Android malware using high-efficient Deep Artificial Neural Network, Future Generation Computer Systems 115, 844-856. – (A2-10) We added the two references (lines 65-67 in related work section). Attachments Attachment Submitted filename: rebuttal_letter.pdf https://doi.org/10.1371/journal.pone.0256187.r002
2 Aug 2021 Decision Letter - Yuchen Qiu, Editor MustaD: Compressing Deep Graph Convolution Network with Multi-Staged Knowledge Distillation PONE-D-21-15818R1 Dear Dr. Kang, We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements. Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication. An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org. If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org. Kind regards, Yuchen Qiu, Ph.D. Academic Editor PLOS ONE Additional Editor Comments (optional): Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation. ******** 2. Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented. Reviewer #2: Yes ****** 3. Has the statistical analysis been performed appropriately and rigorously? Reviewer #2: Yes ****** 4. Have the authors made all data underlying the findings in their manuscript fully available? The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified. Reviewer #2: Yes ****** 5. Is the manuscript presented in an intelligible fashion and written in standard English? PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here. Reviewer #2: Yes ****** 6. Review Comments to the Author Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters) Reviewer #2: I would like to accept this paper now. ****** 7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review?** For information about this choice, including consent withdrawal, please see our Privacy Policy. Reviewer #2: No https://doi.org/10.1371/journal.pone.0256187.r003
Formally Accepted
5 Aug 2021 Acceptance Letter - Yuchen Qiu, Editor PONE-D-21-15818R1 Compressing Deep Graph Convolution Network with Multi-Staged Knowledge Distillation Dear Dr. Kang: I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department. If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org. If we can help with anything else, please email us at plosone@plos.org. Thank you for submitting your work to PLOS ONE and supporting open access. Kind regards, PLOS ONE Editorial Office Staff on behalf of Dr. Yuchen Qiu Academic Editor PLOS ONE https://doi.org/10.1371/journal.pone.0256187.r004

Open letter on the publication of peer review reports

PLOS recognizes the benefits of transparency in the peer review process. Therefore, we enable the publication of all of the content of peer review and author responses alongside final, published articles. Reviewers remain anonymous, unless they choose to reveal their names.

We encourage other journals to join us in this initiative. We hope that our action inspires the community, including researchers, research funders, and research institutions, to recognize the benefits of published peer review reports for all parts of the research system.

Learn more at ASAPbio .