MobileNet-HeX: Heterogeneous Ensemble of MobileNet eXperts for Efficient and Scalable Vision Model Optimization
Efficient and accurate vision models are essential for real-world applications such as medical imaging and deepfake detection, where both performance and computational efficiency are critical. While recent vision models achieve high accuracy, they often come with the trade-off of increased size and...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Big Data and Cognitive Computing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2504-2289/9/1/2 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832589122528608256 |
|---|---|
| author | Emmanuel Pintelas Ioannis E. Livieris Vasilis Tampakas Panagiotis Pintelas |
| author_facet | Emmanuel Pintelas Ioannis E. Livieris Vasilis Tampakas Panagiotis Pintelas |
| author_sort | Emmanuel Pintelas |
| collection | DOAJ |
| description | Efficient and accurate vision models are essential for real-world applications such as medical imaging and deepfake detection, where both performance and computational efficiency are critical. While recent vision models achieve high accuracy, they often come with the trade-off of increased size and computational demands. In this work, we propose MobileNet-HeX, a new ensemble model based on Heterogeneous MobileNet eXperts, designed to achieve top-tier performance while minimizing computational demands in real-world vision tasks. By utilizing a two-step Expand-and-Squeeze mechanism, MobileNet-HeX first expands a MobileNet population through diverse random training setups. It then squeezes the population through pruning, selecting the top-performing models based on heterogeneity and validation performance metrics. Finally, the selected Heterogeneous eXpert MobileNets are combined via sequential quadratic programming to form an efficient super-learner. MobileNet-HeX is benchmarked against state-of-the-art vision models in challenging case studies, such as skin cancer classification and deepfake detection. The results demonstrate that MobileNet-HeX not only surpasses these models in performance but also excels in speed and memory efficiency. By effectively leveraging a diverse set of MobileNet eXperts, we experimentally show that small, yet highly optimized, models can outperform even the most powerful vision networks in both accuracy and computational efficiency. |
| format | Article |
| id | doaj-art-736e4cdd59fe4acc912762108eaa367c |
| institution | Kabale University |
| issn | 2504-2289 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Big Data and Cognitive Computing |
| spelling | doaj-art-736e4cdd59fe4acc912762108eaa367c2025-01-24T13:22:31ZengMDPI AGBig Data and Cognitive Computing2504-22892024-12-0191210.3390/bdcc9010002MobileNet-HeX: Heterogeneous Ensemble of MobileNet eXperts for Efficient and Scalable Vision Model OptimizationEmmanuel Pintelas0Ioannis E. Livieris1Vasilis Tampakas2Panagiotis Pintelas3Department of Mathematics, University of Patras, GR 265-00 Patras, GreeceDepartment of Statistics & Insurance Science, University of Piraeus, GR 185-32 Piraeus, GreeceDepartment of Electrical and Computer Engineering, University of Peloponnese, GR 263-34 Patras, GreeceDepartment of Mathematics, University of Patras, GR 265-00 Patras, GreeceEfficient and accurate vision models are essential for real-world applications such as medical imaging and deepfake detection, where both performance and computational efficiency are critical. While recent vision models achieve high accuracy, they often come with the trade-off of increased size and computational demands. In this work, we propose MobileNet-HeX, a new ensemble model based on Heterogeneous MobileNet eXperts, designed to achieve top-tier performance while minimizing computational demands in real-world vision tasks. By utilizing a two-step Expand-and-Squeeze mechanism, MobileNet-HeX first expands a MobileNet population through diverse random training setups. It then squeezes the population through pruning, selecting the top-performing models based on heterogeneity and validation performance metrics. Finally, the selected Heterogeneous eXpert MobileNets are combined via sequential quadratic programming to form an efficient super-learner. MobileNet-HeX is benchmarked against state-of-the-art vision models in challenging case studies, such as skin cancer classification and deepfake detection. The results demonstrate that MobileNet-HeX not only surpasses these models in performance but also excels in speed and memory efficiency. By effectively leveraging a diverse set of MobileNet eXperts, we experimentally show that small, yet highly optimized, models can outperform even the most powerful vision networks in both accuracy and computational efficiency.https://www.mdpi.com/2504-2289/9/1/2ensemble learningMobileNetconvolutional neural networksimage classification |
| spellingShingle | Emmanuel Pintelas Ioannis E. Livieris Vasilis Tampakas Panagiotis Pintelas MobileNet-HeX: Heterogeneous Ensemble of MobileNet eXperts for Efficient and Scalable Vision Model Optimization Big Data and Cognitive Computing ensemble learning MobileNet convolutional neural networks image classification |
| title | MobileNet-HeX: Heterogeneous Ensemble of MobileNet eXperts for Efficient and Scalable Vision Model Optimization |
| title_full | MobileNet-HeX: Heterogeneous Ensemble of MobileNet eXperts for Efficient and Scalable Vision Model Optimization |
| title_fullStr | MobileNet-HeX: Heterogeneous Ensemble of MobileNet eXperts for Efficient and Scalable Vision Model Optimization |
| title_full_unstemmed | MobileNet-HeX: Heterogeneous Ensemble of MobileNet eXperts for Efficient and Scalable Vision Model Optimization |
| title_short | MobileNet-HeX: Heterogeneous Ensemble of MobileNet eXperts for Efficient and Scalable Vision Model Optimization |
| title_sort | mobilenet hex heterogeneous ensemble of mobilenet experts for efficient and scalable vision model optimization |
| topic | ensemble learning MobileNet convolutional neural networks image classification |
| url | https://www.mdpi.com/2504-2289/9/1/2 |
| work_keys_str_mv | AT emmanuelpintelas mobilenethexheterogeneousensembleofmobilenetexpertsforefficientandscalablevisionmodeloptimization AT ioanniselivieris mobilenethexheterogeneousensembleofmobilenetexpertsforefficientandscalablevisionmodeloptimization AT vasilistampakas mobilenethexheterogeneousensembleofmobilenetexpertsforefficientandscalablevisionmodeloptimization AT panagiotispintelas mobilenethexheterogeneousensembleofmobilenetexpertsforefficientandscalablevisionmodeloptimization |