paper-with-me

Papers

Beyond One-Way Pruning: Bidirectional Pruning-Regrowth for Extreme Accuracy-Sparsity Tradeoff

2025-11-11 · Junchen Liu, Yi Sheng arxiv

As a widely adopted model compression technique, model pruning has demonstrated strong effectiveness across various architectures. However, we observe that when sparsity exceeds a certain threshold, both iterative and one-shot pruning methods lead to a steep decline in model performance. This rapid degradation limits the achievable compression ratio and prevents models from meeting the stringent size constraints required by certain hardware platforms, rendering them inoperable. To overcome this limitation, we propose a bidirectional pruning-regrowth strategy. Starting from an extremely compressed network that satisfies hardware constraints, the method selectively regenerates critical connections to recover lost performance, effectively mitigating the sharp accuracy drop commonly observed under high sparsity conditions.

📄 PDF Abstract BibTeX arXiv:2511.11675

Code (0)

등록된 구현이 없습니다.

Tasks

Model Compression

Similar Papers 제목 키워드 기반

Pruning of Deep Spiking Neural Networks through Gradient Rewiring

2021-05-11 · Yanqi Chen, Zhaofei Yu, Wei Fang, Tiejun Huang 외

Spiking Neural Networks (SNNs) have been attached great importance due to their biological plausibility and high energy-efficiency on neuromorphic chips. As these chips are usually resource-constrained, the compression o…

EntryPrune: Neural Network Feature Selection using First Impressions

2024-10-03 · Felix Zimmer, Patrik Okanovic, Torsten Hoefler

There is an ongoing effort to develop feature selection algorithms to improve interpretability, reduce computational resources, and minimize overfitting in predictive models. Neural networks stand out as architectures on…

feature selection

Comprehensive Graph Gradual Pruning for Sparse Training in Graph Neural Networks

2022-07-18 · Chuang Liu, Xueqi Ma, Yibing Zhan, Liang Ding 외

Graph Neural Networks (GNNs) tend to suffer from high computation costs due to the exponentially increasing scale of graph data and the number of model parameters, which restricts their utility in practical applications.…

Node Classification

CA-AFP: Cluster-Aware Adaptive Federated Pruning

2026-03-02 · Om Govind Jha, Harsh Shukla, Haroon R. Lone arxiv

Federated Learning (FL) faces major challenges in real-world deployments due to statistical heterogeneity across clients and system heterogeneity arising from resource-constrained devices. While clustering-based approach…

Human Activity RecognitionFederated Learning

A procedure for automated tree pruning suggestion using LiDAR scans of fruit trees

2021-02-07 · Fredrik Westling, James Underwood, Mitch Bryson

In fruit tree growth, pruning is an important management practice for preventing overcrowding, improving canopy access to light and promoting regrowth. Due to the slow nature of agriculture, decisions in pruning are typi…

Decision MakingManagement