Abstract
Over the past decades, deep learning (DL) systems have achieved tremendous success and gained great popularity in various applications, such as intelligent machines, image processing, speech processing, and medical diagnostics. Deep neural networks are the key driving force behind its recent success, but still seem to be a magic black box lacking interpretability and understanding. This brings up many open safety and security issues with enormous and urgent demands on rigorous methodologies and engineering practice for quality enhancement. A plethora of studies have shown that state-of-the-art DL systems suffer from defects and vulnerabilities that can lead to severe loss and tragedies, especially when applied to real-world safety-critical applications. In this paper, we perform a large-scale study and construct a paper repository of 223 relevant works to the quality assurance, security, and interpretation of deep learning. Based on this, we, from a software quality assurance perspective, pinpoint challenges and future opportunities to facilitate drawing the attention of the software engineering community towards addressing the pressing industrial demand of secure intelligent systems.
| Original language | English |
|---|---|
| Title of host publication | Formal Methods and Software Engineering - 21st International Conference on Formal Engineering Methods, ICFEM 2019, Proceedings |
| Editors | Yamine Ait-Ameur, Shengchao Qin |
| Publisher | Springer |
| Pages | 3-15 |
| Number of pages | 13 |
| ISBN (Print) | 9783030324087 |
| DOIs | |
| Publication status | Published - Jan 1 2019 |
| Event | 21st International Conference on Formal Engineering Methods, ICFEM 2019 - Shenzhen, China Duration: Nov 5 2019 → Nov 9 2019 |
Publication series
| Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
|---|---|
| Volume | 11852 LNCS |
| ISSN (Print) | 0302-9743 |
| ISSN (Electronic) | 1611-3349 |
Conference
| Conference | 21st International Conference on Formal Engineering Methods, ICFEM 2019 |
|---|---|
| Country | China |
| City | Shenzhen |
| Period | 11/5/19 → 11/9/19 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Theoretical Computer Science
- Computer Science(all)
Cite this
Secure Deep Learning Engineering : A Road Towards Quality Assurance of Intelligent Systems. / Liu, Yang; Ma, Lei; Zhao, Jianjun.
Formal Methods and Software Engineering - 21st International Conference on Formal Engineering Methods, ICFEM 2019, Proceedings. ed. / Yamine Ait-Ameur; Shengchao Qin. Springer, 2019. p. 3-15 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11852 LNCS).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Secure Deep Learning Engineering
T2 - A Road Towards Quality Assurance of Intelligent Systems
AU - Liu, Yang
AU - Ma, Lei
AU - Zhao, Jianjun
PY - 2019/1/1
Y1 - 2019/1/1
N2 - Over the past decades, deep learning (DL) systems have achieved tremendous success and gained great popularity in various applications, such as intelligent machines, image processing, speech processing, and medical diagnostics. Deep neural networks are the key driving force behind its recent success, but still seem to be a magic black box lacking interpretability and understanding. This brings up many open safety and security issues with enormous and urgent demands on rigorous methodologies and engineering practice for quality enhancement. A plethora of studies have shown that state-of-the-art DL systems suffer from defects and vulnerabilities that can lead to severe loss and tragedies, especially when applied to real-world safety-critical applications. In this paper, we perform a large-scale study and construct a paper repository of 223 relevant works to the quality assurance, security, and interpretation of deep learning. Based on this, we, from a software quality assurance perspective, pinpoint challenges and future opportunities to facilitate drawing the attention of the software engineering community towards addressing the pressing industrial demand of secure intelligent systems.
AB - Over the past decades, deep learning (DL) systems have achieved tremendous success and gained great popularity in various applications, such as intelligent machines, image processing, speech processing, and medical diagnostics. Deep neural networks are the key driving force behind its recent success, but still seem to be a magic black box lacking interpretability and understanding. This brings up many open safety and security issues with enormous and urgent demands on rigorous methodologies and engineering practice for quality enhancement. A plethora of studies have shown that state-of-the-art DL systems suffer from defects and vulnerabilities that can lead to severe loss and tragedies, especially when applied to real-world safety-critical applications. In this paper, we perform a large-scale study and construct a paper repository of 223 relevant works to the quality assurance, security, and interpretation of deep learning. Based on this, we, from a software quality assurance perspective, pinpoint challenges and future opportunities to facilitate drawing the attention of the software engineering community towards addressing the pressing industrial demand of secure intelligent systems.
UR - http://www.scopus.com/inward/record.url?scp=85076169318&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85076169318&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-32409-4_1
DO - 10.1007/978-3-030-32409-4_1
M3 - Conference contribution
AN - SCOPUS:85076169318
SN - 9783030324087
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 3
EP - 15
BT - Formal Methods and Software Engineering - 21st International Conference on Formal Engineering Methods, ICFEM 2019, Proceedings
A2 - Ait-Ameur, Yamine
A2 - Qin, Shengchao
PB - Springer
ER -