paper-with-me

홈 › Papers

Sparsifying the Update Step in Graph Neural Networks

2021-09-02 · Johannes F. Lutzeyer, Changmin Wu, Michalis Vazirgiannis

Message-Passing Neural Networks (MPNNs), the most prominent Graph Neural Network (GNN) framework, celebrate much success in the analysis of graph-structured data. Concurrently, the sparsification of Neural Network models attracts a great amount of academic and industrial interest. In this paper we conduct a structured, empirical study of the effect of sparsification on the trainable part of MPNNs known as the Update step. To this end, we design a series of models to successively sparsify the linear transform in the Update step. Specifically, we propose the ExpanderGNN model with a tuneable sparsification rate and the Activation-Only GNN, which has no linear transform in the Update step. In agreement with a growing trend in the literature the sparsification paradigm is changed by initialising sparse neural network architectures rather than expensively sparsifying already trained architectures. Our novel benchmark models enable a better understanding of the influence of the Update step on model performance and outperform existing simplified benchmark models such as the Simple Graph Convolution. The ExpanderGNNs, and in some cases the Activation-Only models, achieve performance on par with their vanilla counterparts on several downstream tasks, while containing significantly fewer trainable parameters. Our code is publicly available at: https://github.com/ChangminWu/ExpanderGNN.

📄 PDF Abstract BibTeX arXiv:2109.00909

Code (1)

changminwu/expandergnn 공식 구현 pytorch

Tasks

Graph Neural Network

Methods 이 논문이 사용한 방법론

Graph Neural Network 설명 없음
Convolution A convolution is a type of matrix operation, consisting of a kernel, a small matrix of weights, that slides over input data performing element-wise multiplication with the…

Similar Papers 제목 키워드 기반

$\ell_0$ Sparsifying Transform Learning with Efficient Optimal Updates and Convergence Guarantees

2015-01-13 · Saiprasad Ravishankar, Yoram Bresler

Many applications in signal processing benefit from the sparsity of signals in a certain transform domain or dictionary. Synthesis sparsifying dictionaries that are directly adapted to data have been popular in applicati…

DenoisingImage DenoisingImage Reconstruction

Low Dose CT Image Reconstruction With Learned Sparsifying Transform

2017-07-10 · Xuehang Zheng, Zening Lu, Saiprasad Ravishankar, Yong Long 외

A major challenge in computed tomography (CT) is to reduce X-ray dose to a low or even ultra-low level while maintaining the high quality of reconstructed images. We propose a new method for CT reconstruction that combin…

Computed Tomography (CT)CT ReconstructionImage Reconstruction

Analysing the Update step in Graph Neural Networks via Sparsification

2021-01-01 · Changmin Wu, Johannes F. Lutzeyer, Michalis Vazirgiannis

In recent years, Message-Passing Neural Networks (MPNNs), the most prominent Graph Neural Network (GNN) framework, have celebrated much success in the analysis of graph-structured data. In MPNNs the computations are spli…

Graph Neural Network

FRIST - Flipping and Rotation Invariant Sparsifying Transform Learning and Applications

2015-11-19 · Bihan Wen, Saiprasad Ravishankar, Yoram Bresler

Features based on sparse representation, especially using the synthesis dictionary model, have been heavily exploited in signal processing and computer vision. However, synthesis dictionary learning typically involves NP…

compressed sensingDenoisingDictionary LearningImage Reconstruction

Learned Multi-layer Residual Sparsifying Transform Model for Low-dose CT Reconstruction

2020-05-08 · Xikai Yang, Xuehang Zheng, Yong Long, Saiprasad Ravishankar

Signal models based on sparse representation have received considerable attention in recent years. Compared to synthesis dictionary learning, sparsifying transform learning involves highly efficient sparse coding and ope…

CT ReconstructionDictionary Learning