Secure deep learning for distributed data against maliciouscentral server
In this paper, we propose a secure system for performing deep learning with distributed trainers connected to a central parameter server. Our system has the following two distinct features: (1) the distributed trainers can detect malicious activities in the server; (2) the distributed trainers can p...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2022-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9342767/?tool=EBI |
| _version_ | 1811221404909043712 |
|---|---|
| author | Le Trieu Phong |
| author_facet | Le Trieu Phong |
| author_sort | Le Trieu Phong |
| collection | DOAJ |
| description | In this paper, we propose a secure system for performing deep learning with distributed trainers connected to a central parameter server. Our system has the following two distinct features: (1) the distributed trainers can detect malicious activities in the server; (2) the distributed trainers can perform both vertical and horizontal neural network training. In the experiments, we apply our system to medical data including magnetic resonance and X-ray images and obtain approximate or even better area-under-the-curve scores when compared to the existing scores. |
| first_indexed | 2024-04-12T07:58:42Z |
| format | Article |
| id | doaj.art-600f1c549e7d4e78b2261fa7c6a2f487 |
| institution | Directory Open Access Journal |
| issn | 1932-6203 |
| language | English |
| last_indexed | 2024-04-12T07:58:42Z |
| publishDate | 2022-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj.art-600f1c549e7d4e78b2261fa7c6a2f4872022-12-22T03:41:23ZengPublic Library of Science (PLoS)PLoS ONE1932-62032022-01-01178Secure deep learning for distributed data against maliciouscentral serverLe Trieu PhongIn this paper, we propose a secure system for performing deep learning with distributed trainers connected to a central parameter server. Our system has the following two distinct features: (1) the distributed trainers can detect malicious activities in the server; (2) the distributed trainers can perform both vertical and horizontal neural network training. In the experiments, we apply our system to medical data including magnetic resonance and X-ray images and obtain approximate or even better area-under-the-curve scores when compared to the existing scores.https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9342767/?tool=EBI |
| spellingShingle | Le Trieu Phong Secure deep learning for distributed data against maliciouscentral server PLoS ONE |
| title | Secure deep learning for distributed data against maliciouscentral server |
| title_full | Secure deep learning for distributed data against maliciouscentral server |
| title_fullStr | Secure deep learning for distributed data against maliciouscentral server |
| title_full_unstemmed | Secure deep learning for distributed data against maliciouscentral server |
| title_short | Secure deep learning for distributed data against maliciouscentral server |
| title_sort | secure deep learning for distributed data against maliciouscentral server |
| url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9342767/?tool=EBI |
| work_keys_str_mv | AT letrieuphong securedeeplearningfordistributeddataagainstmaliciouscentralserver |