Mixture of Experts for Node Classification

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shi, Yu, Wang, Yiqi, Lang, WeiXuan, Zhang, Jiaxin, Dong, Pan, Li, Aiping
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916776578646016
author Shi, Yu
Wang, Yiqi
Lang, WeiXuan
Zhang, Jiaxin
Dong, Pan
Li, Aiping
author_facet Shi, Yu
Wang, Yiqi
Lang, WeiXuan
Zhang, Jiaxin
Dong, Pan
Li, Aiping
contents Nodes in the real-world graphs exhibit diverse patterns in numerous aspects, such as degree and homophily. However, most existent node predictors fail to capture a wide range of node patterns or to make predictions based on distinct node patterns, resulting in unsatisfactory classification performance. In this paper, we reveal that different node predictors are good at handling nodes with specific patterns and only apply one node predictor uniformly could lead to suboptimal result. To mitigate this gap, we propose a mixture of experts framework, MoE-NP, for node classification. Specifically, MoE-NP combines a mixture of node predictors and strategically selects models based on node patterns. Experimental results from a range of real-world datasets demonstrate significant performance improvements from MoE-NP.
format Preprint
id arxiv_https___arxiv_org_abs_2412_00418
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Mixture of Experts for Node Classification
Shi, Yu
Wang, Yiqi
Lang, WeiXuan
Zhang, Jiaxin
Dong, Pan
Li, Aiping
Social and Information Networks
Artificial Intelligence
Nodes in the real-world graphs exhibit diverse patterns in numerous aspects, such as degree and homophily. However, most existent node predictors fail to capture a wide range of node patterns or to make predictions based on distinct node patterns, resulting in unsatisfactory classification performance. In this paper, we reveal that different node predictors are good at handling nodes with specific patterns and only apply one node predictor uniformly could lead to suboptimal result. To mitigate this gap, we propose a mixture of experts framework, MoE-NP, for node classification. Specifically, MoE-NP combines a mixture of node predictors and strategically selects models based on node patterns. Experimental results from a range of real-world datasets demonstrate significant performance improvements from MoE-NP.
title Mixture of Experts for Node Classification
topic Social and Information Networks
Artificial Intelligence
url https://arxiv.org/abs/2412.00418