RMNv2: Reduced Mobilenet V2 for CIFAR10

dc.contributor.authorAyi, Maneesh
dc.contributor.authorEl-Sharkawy, Mohamed
dc.contributor.departmentElectrical and Computer Engineering, School of Engineering and Technologyen_US
dc.date.accessioned2021-03-03T20:45:40Z
dc.date.available2021-03-03T20:45:40Z
dc.date.issued2020-01
dc.description.abstractIn this paper, we developed a new architecture called Reduced Mobilenet V2 (RMNv2) for CIFAR10 dataset. The baseline architecture of our network is Mobilenet V2. RMNv2 is architecturally modified version of Mobilenet V2. The proposed model has a total number of parameters of 1.06M which is 52.2% lesser than the baseline model. The overall accuracy of RMNv2 for CIFAR10 dataset is 92.4% which is 1.9% lesser than the baseline model. The architectural modifications involve heterogeneous kernel-based convolutions, mish activation, etc. Also, we include a data augmentation technique called AutoAugment that contributes to increasing accuracy of our model. This architectural modification makes the model suitable for resource-constrained devices like embedded devices, mobile devices deployment for real-time applications like autonomous vehicles, object recognition, etc.en_US
dc.eprint.versionAuthor's manuscripten_US
dc.identifier.citationAyi, M., & El-Sharkawy, M. (2020). RMNv2: Reduced Mobilenet V2 for CIFAR10. 2020 10th Annual Computing and Communication Workshop and Conference (CCWC), 0287–0292. https://doi.org/10.1109/CCWC47524.2020.9031131en_US
dc.identifier.urihttps://hdl.handle.net/1805/25307
dc.language.isoenen_US
dc.publisherIEEEen_US
dc.relation.isversionof10.1109/CCWC47524.2020.9031131en_US
dc.relation.journal2020 10th Annual Computing and Communication Workshop and Conferenceen_US
dc.rightsPublisher Policyen_US
dc.sourceAuthoren_US
dc.subjectMobilenetV2 architectureen_US
dc.subjectconvolution neural networksen_US
dc.subjectdeep neural networksen_US
dc.titleRMNv2: Reduced Mobilenet V2 for CIFAR10en_US
dc.typeConference proceedingsen_US
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Ayi2030RMNv2.pdf
Size:
1.19 MB
Format:
Adobe Portable Document Format
Description:
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.99 KB
Format:
Item-specific license agreed upon to submission
Description: