Final published version
Research output: Contribution to Journal/Magazine › Journal article › peer-review
Research output: Contribution to Journal/Magazine › Journal article › peer-review
}
TY - JOUR
T1 - A survey of safety and trustworthiness of deep neural networks
T2 - Verification, testing, adversarial attack and defence, and interpretability
AU - Huang, X.
AU - Kroening, D.
AU - Ruan, W.
AU - Sharp, J.
AU - Sun, Y.
AU - Thamo, E.
AU - Wu, M.
AU - Yi, X.
PY - 2020/8/1
Y1 - 2020/8/1
N2 - In the past few years, significant progress has been made on deep neural networks (DNNs) in achieving human-level performance on several long-standing tasks. With the broader deployment of DNNs on various applications, the concerns over their safety and trustworthiness have been raised in public, especially after the widely reported fatal incidents involving self-driving cars. Research to address these concerns is particularly active, with a significant number of papers released in the past few years. This survey paper conducts a review of the current research effort into making DNNs safe and trustworthy, by focusing on four aspects: verification, testing, adversarial attack and defence, and interpretability. In total, we survey 202 papers, most of which were published after 2017.
AB - In the past few years, significant progress has been made on deep neural networks (DNNs) in achieving human-level performance on several long-standing tasks. With the broader deployment of DNNs on various applications, the concerns over their safety and trustworthiness have been raised in public, especially after the widely reported fatal incidents involving self-driving cars. Research to address these concerns is particularly active, with a significant number of papers released in the past few years. This survey paper conducts a review of the current research effort into making DNNs safe and trustworthy, by focusing on four aspects: verification, testing, adversarial attack and defence, and interpretability. In total, we survey 202 papers, most of which were published after 2017.
KW - Deep neural networks
KW - Safety testing
KW - Surveys
KW - Fatal incidents
KW - Human-level performance
KW - Interpretability
KW - Research efforts
KW - Standing tasks
KW - Neural networks
U2 - 10.1016/j.cosrev.2020.100270
DO - 10.1016/j.cosrev.2020.100270
M3 - Journal article
VL - 37
JO - Computer Science Review
JF - Computer Science Review
SN - 1574-0137
M1 - 100270
ER -