Saved in:
Bibliographic Details
Main Authors: Wang, Xuan, Pang, Zeshan, Lu, Yuliang, Yan, Xuehu
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2402.17486
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917599375261696
author Wang, Xuan
Pang, Zeshan
Lu, Yuliang
Yan, Xuehu
author_facet Wang, Xuan
Pang, Zeshan
Lu, Yuliang
Yan, Xuehu
contents To provide a foundation for the research of deep learning models, the construction of model pool is an essential step. This paper proposes a Training-Free and Efficient Model Generation and Enhancement Scheme (MGE). This scheme primarily considers two aspects during the model generation process: the distribution of model parameters and model performance. Experiments result shows that generated models are comparable to models obtained through normal training, and even superior in some cases. Moreover, the time consumed in generating models accounts for only 1\% of the time required for normal model training. More importantly, with the enhancement of Evolution-MGE, generated models exhibits competitive generalization ability in few-shot tasks. And the behavioral dissimilarity of generated models has the potential of adversarial defense.
format Preprint
id arxiv_https___arxiv_org_abs_2402_17486
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle MGE: A Training-Free and Efficient Model Generation and Enhancement Scheme
Wang, Xuan
Pang, Zeshan
Lu, Yuliang
Yan, Xuehu
Computer Vision and Pattern Recognition
To provide a foundation for the research of deep learning models, the construction of model pool is an essential step. This paper proposes a Training-Free and Efficient Model Generation and Enhancement Scheme (MGE). This scheme primarily considers two aspects during the model generation process: the distribution of model parameters and model performance. Experiments result shows that generated models are comparable to models obtained through normal training, and even superior in some cases. Moreover, the time consumed in generating models accounts for only 1\% of the time required for normal model training. More importantly, with the enhancement of Evolution-MGE, generated models exhibits competitive generalization ability in few-shot tasks. And the behavioral dissimilarity of generated models has the potential of adversarial defense.
title MGE: A Training-Free and Efficient Model Generation and Enhancement Scheme
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2402.17486