Classic GNNs are Strong Baselines: Reassessing GNNs for Node Classification

计算机科学 生物
作者
Yuankai Luo,Lei Shi,Xiao-Ming Wu
出处
期刊:Cornell University - arXiv
标识
DOI:10.48550/arxiv.2406.08993
摘要

Graph Transformers (GTs) have recently emerged as popular alternatives to traditional message-passing Graph Neural Networks (GNNs), due to their theoretically superior expressiveness and impressive performance reported on standard node classification benchmarks, often significantly outperforming GNNs. In this paper, we conduct a thorough empirical analysis to reevaluate the performance of three classic GNN models (GCN, GAT, and GraphSAGE) against GTs. Our findings suggest that the previously reported superiority of GTs may have been overstated due to suboptimal hyperparameter configurations in GNNs. Remarkably, with slight hyperparameter tuning, these classic GNN models achieve state-of-the-art performance, matching or even exceeding that of recent GTs across 17 out of the 18 diverse datasets examined. Additionally, we conduct detailed ablation studies to investigate the influence of various GNN configurations, such as normalization, dropout, residual connections, network depth, and jumping knowledge mode, on node classification performance. Our study aims to promote a higher standard of empirical rigor in the field of graph machine learning, encouraging more accurate comparisons and evaluations of model capabilities.

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
zz完成签到,获得积分10
刚刚
好玉完成签到,获得积分10
1秒前
2秒前
落寞冬日完成签到,获得积分10
2秒前
Ju发布了新的文献求助10
2秒前
cdercder应助溪泉采纳,获得10
2秒前
3秒前
李雨完成签到,获得积分10
3秒前
li发布了新的文献求助10
3秒前
3秒前
4秒前
科研通AI2S应助直率冷之采纳,获得30
4秒前
Jasper应助hope采纳,获得10
4秒前
Sssssss完成签到,获得积分10
4秒前
殷勤的觅松完成签到,获得积分10
4秒前
眯眯眼的南风完成签到,获得积分10
4秒前
Sh1zuru完成签到,获得积分10
5秒前
凉夏完成签到,获得积分10
5秒前
sxiao18发布了新的文献求助10
5秒前
赘婿应助asn采纳,获得10
5秒前
自由的灵萱完成签到 ,获得积分10
5秒前
llll发布了新的文献求助50
5秒前
Jasper应助5U采纳,获得10
5秒前
5秒前
6秒前
hm应助科研通管家采纳,获得10
6秒前
6秒前
四季如春完成签到,获得积分10
7秒前
bao完成签到,获得积分10
7秒前
李健应助科研通管家采纳,获得10
7秒前
arniu2008应助科研通管家采纳,获得150
7秒前
xjcy应助科研通管家采纳,获得10
7秒前
7秒前
Ava应助科研通管家采纳,获得10
7秒前
汉堡包应助科研通管家采纳,获得10
7秒前
xjcy应助科研通管家采纳,获得10
7秒前
充电宝应助科研通管家采纳,获得10
7秒前
蓝天应助科研通管家采纳,获得10
8秒前
喜来乐完成签到,获得积分10
8秒前
cdercder应助科研通管家采纳,获得10
8秒前
高分求助中
Annie Ernaux: De la perte au corps glorieux 600
Petrology and Plate Tectonics,2025 500
Optical Coating Design with the Essential Macleod 400
A revision of Limenitis helmanni and its related species (Nymphalidae) from Central and South China 400
Moore's Clinically Oriented Anatomy 10th Edition 400
Direct and Iterative Linear System Solvers 400
Cardiopulmonary Bypass and Mechanical Support: Principles and Practice, Fifth Edition 400
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 工程类 有机化学 化学工程 生物化学 计算机科学 物理 内科学 复合材料 催化作用 物理化学 光电子学 电极 细胞生物学 基因 无机化学
热门帖子
关注 科研通微信公众号,转发送积分 6784665
求助须知:如何正确求助?哪些是违规求助? 8506780
关于积分的说明 18117187
捐赠科研通 6090095
什么是DOI,文献DOI怎么找? 3019760
邀请新用户注册赠送积分活动 1996736
关于科研通互助平台的介绍 1982883