Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

multi-stage GNN

A multi-stage graph neural network is a graph neural network architecture that processes and encodes graph-structured data across multiple hierarchical levels or sequential operational phases. Rather than relying on a flat embedding approach across all nodes simultaneously, this framework decomposes complex graphs into distinct stages, such as local subgraphs or repetitive motifs at lower levels and broader topological configurations at higher levels. Through phased message passing and pooling mechanisms across these stages, a multi-stage graph neural network effectively aggregates both fine-grained local features and high-level structural patterns into comprehensive, multi-scale graph embeddings.

1 item

Topology-Aware Network Pruning using Multi-stage Graph Embedding and Reinforcement Learning

Topology-Aware Network Pruning using Multi-stage Graph Embedding and Reinforcement Learning

Sixing Yu, Arya Mazaheri, Ali Jannesari

OrganizationsIowa State UniversityTechnische Universität Darmstadt

Why you should read this

Proposes a multi-stage graph embedding framework that captures hierarchical neural network topologies to automatically learn optimal, FLOP-constrained pruning policies via reinforcement learning across diverse architectures.

Model compression is an essential technique for deploying deep neural networks (DNNs) on power and memory-constrained resources. However, existing model-compression methods often rely on human expertise and focus on parameters’ local importance, ignoring the rich topology information within DNNs. In this paper, we propose a novel multi-stage graph embedding technique based on graph neural networks (GNNs) to identify DNN topologies and use reinforcement learning (RL) to find a suitable compression policy. We performed resource-constrained (i.e., FLOPs) channel pruning and compared our approach with state-of-the-art model compression methods. We evaluated our method on various models from typical to mobile-friendly networks, such as ResNet family, VGG-16, MobileNet-v1/v2, and ShuffleNet. Results show that our method can achieve higher compression ratios with a minimal fine-tuning cost yet yields outstanding and competitive performance. The code is open-sourced at https://github.com/yusx-swapp/GNN-RL-Model-Compression.

Added

2026-10-03