PruneFuse: Efficient Data Selection via Weight Pruning and Network Fusion

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kousar, Humaira, Bhatti, Hasnain Irshad, Moon, Jaekyun
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911548006465536
author Kousar, Humaira
Bhatti, Hasnain Irshad
Moon, Jaekyun
author_facet Kousar, Humaira
Bhatti, Hasnain Irshad
Moon, Jaekyun
contents Efficient data selection is crucial for enhancing the training efficiency of deep neural networks and minimizing annotation requirements. Traditional methods often face high computational costs, limiting their scalability and practical use. We introduce PruneFuse, a novel strategy that leverages pruned networks for data selection and later fuses them with the original network to optimize training. PruneFuse operates in two stages: First, it applies structured pruning to create a smaller pruned network that, due to its structural coherence with the original network, is well-suited for the data selection task. This small network is then trained and selects the most informative samples from the dataset. Second, the trained pruned network is seamlessly fused with the original network. This integration leverages the insights gained during the training of the pruned network to facilitate the learning process of the fused network while leaving room for the network to discover more robust solutions. Extensive experimentation on various datasets demonstrates that PruneFuse significantly reduces computational costs for data selection, achieves better performance than baselines, and accelerates the overall training process.
format Preprint
id arxiv_https___arxiv_org_abs_2603_26138
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle PruneFuse: Efficient Data Selection via Weight Pruning and Network Fusion
Kousar, Humaira
Bhatti, Hasnain Irshad
Moon, Jaekyun
Machine Learning
Computer Vision and Pattern Recognition
Efficient data selection is crucial for enhancing the training efficiency of deep neural networks and minimizing annotation requirements. Traditional methods often face high computational costs, limiting their scalability and practical use. We introduce PruneFuse, a novel strategy that leverages pruned networks for data selection and later fuses them with the original network to optimize training. PruneFuse operates in two stages: First, it applies structured pruning to create a smaller pruned network that, due to its structural coherence with the original network, is well-suited for the data selection task. This small network is then trained and selects the most informative samples from the dataset. Second, the trained pruned network is seamlessly fused with the original network. This integration leverages the insights gained during the training of the pruned network to facilitate the learning process of the fused network while leaving room for the network to discover more robust solutions. Extensive experimentation on various datasets demonstrates that PruneFuse significantly reduces computational costs for data selection, achieves better performance than baselines, and accelerates the overall training process.
title PruneFuse: Efficient Data Selection via Weight Pruning and Network Fusion
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.26138