Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wei, Timothy, Peng, Hsien Xin, Xu, Elaine, Zhao, Bryan, Ding, Lei, Yang, Diji
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2410.12165
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866929551401025536
author Wei, Timothy
Peng, Hsien Xin
Xu, Elaine
Zhao, Bryan
Ding, Lei
Yang, Diji
author_facet Wei, Timothy
Peng, Hsien Xin
Xu, Elaine
Zhao, Bryan
Ding, Lei
Yang, Diji
contents As Artificial Intelligence models, such as Large Video-Language models (VLMs), grow in size, their deployment in real-world applications becomes increasingly challenging due to hardware limitations and computational costs. To address this, we design a hybrid edge-cloud solution that leverages the efficiency of smaller models for local processing while deferring to larger, more accurate cloud-based models when necessary. Specifically, we propose a novel unsupervised data generation method, Dual-Model Distillation (DMD), to train a lightweight switcher model that can predict when the edge model's output is uncertain and selectively offload inference to the large model in the cloud. Experimental results on the action classification task show that our framework not only requires less computational overhead, but also improves accuracy compared to using a large model alone. Our framework provides a scalable and adaptable solution for action classification in resource-constrained environments, with potential applications beyond healthcare. Noteworthy, while DMD-generated data is used for optimizing performance and resource usage in our pipeline, we expect the concept of DMD to further support future research on knowledge alignment across multiple models.
format Preprint
id arxiv_https___arxiv_org_abs_2410_12165
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Dual-Model Distillation for Efficient Action Classification with Hybrid Edge-Cloud Solution
Wei, Timothy
Peng, Hsien Xin
Xu, Elaine
Zhao, Bryan
Ding, Lei
Yang, Diji
Computer Vision and Pattern Recognition
Artificial Intelligence
As Artificial Intelligence models, such as Large Video-Language models (VLMs), grow in size, their deployment in real-world applications becomes increasingly challenging due to hardware limitations and computational costs. To address this, we design a hybrid edge-cloud solution that leverages the efficiency of smaller models for local processing while deferring to larger, more accurate cloud-based models when necessary. Specifically, we propose a novel unsupervised data generation method, Dual-Model Distillation (DMD), to train a lightweight switcher model that can predict when the edge model's output is uncertain and selectively offload inference to the large model in the cloud. Experimental results on the action classification task show that our framework not only requires less computational overhead, but also improves accuracy compared to using a large model alone. Our framework provides a scalable and adaptable solution for action classification in resource-constrained environments, with potential applications beyond healthcare. Noteworthy, while DMD-generated data is used for optimizing performance and resource usage in our pipeline, we expect the concept of DMD to further support future research on knowledge alignment across multiple models.
title Dual-Model Distillation for Efficient Action Classification with Hybrid Edge-Cloud Solution
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2410.12165