2021
DOI: 10.1007/s00521-021-05960-5
|View full text |Cite
|
Sign up to set email alerts
|

Advanced metaheuristic optimization techniques in applications of deep neural networks: a review

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
2

Citation Types

0
34
0
3

Year Published

2021
2021
2024
2024

Publication Types

Select...
6
2

Relationship

1
7

Authors

Journals

citations
Cited by 95 publications
(37 citation statements)
references
References 112 publications
0
34
0
3
Order By: Relevance
“…The time is represented similarly to strategy 1. Thereby, we obtain the equation of the trajectory, as seen in Equation (5).…”
Section: Cos( ) Cos( )mentioning
confidence: 99%
See 2 more Smart Citations
“…The time is represented similarly to strategy 1. Thereby, we obtain the equation of the trajectory, as seen in Equation (5).…”
Section: Cos( ) Cos( )mentioning
confidence: 99%
“…The time is represented similarly to strategy 1. Thereby, we obtain the equation of the trajectory, as seen in Equation (5). The equations of projectile motion, which result from the composition of a uniform motion along the X-axis and a uniformly accelerated motion along the Y-axis, are as follows:…”
Section: Cos( ) Cos( )mentioning
confidence: 99%
See 1 more Smart Citation
“…Researchers used dimensional reduction to eliminate the unimportant and redundant data, which maps the original high-dimensional space data into new lower-dimensional space data [7,8]. Reducing dimensional also allows to imagine and reflect the data and can increase the application's output [9].…”
Section: Introductionmentioning
confidence: 99%
“…Then, CNN was used to extract features of the incomplete data through the compression layer and gradually extended these features in the reconstruction layer to recover the missing data. Abd Elaziz [22,23] proposed an advanced metaheuristic optimization, called aquila optimizer, which is powerful in tackling missing data. In the above algorithms, the BiGRU model is a data-driven method that does not need many assumptions and can automatically adjust its parameters according to the data.…”
Section: Introductionmentioning
confidence: 99%