根据节点难易度动态分配专家资源，提升图分类精度与效率
Learning How Much to Think: Difficulty-Aware Dynamic MoEs for Graph Node Classification
- 用预测熵衡量节点难度，动态调整专家使用数量
- 在异质图上最高提升7.92%准确率，大图节省73%内存
- 适合需要高效高精度图神经网络的科研与工程场景
混合专家（MoE）架构为图神经网络在节点分类任务中提供了可扩展路径，但通常依赖静态且僵化的路由策略，在所有节点上采用统一的专家预算或粗粒度开关，忽略了节点判别难度的差异，导致难节点欠拟合、易节点冗余计算。为此，我们提出D2MoE框架，将焦点从静态专家选择转向节点级资源分配。通过预测熵作为难度的实时代理，D2MoE采用难度驱动的top-p路由机制，自适应地将专家资源集中于难节点，同时降低易节点的开销，实现节点分类中连续且精细的专家预算调节。在13个基准测试上，D2MoE表现一致领先，异质图上最高超越主流基线7.92%准确率；在大规模图上，相比最佳性能的Graph MoE，内存消耗减少高达73.07%，训练时间缩短46.53%，验证了其卓越效率。
原文摘要 · Abstract (English)
Mixture-of-Experts (MoE) architectures offer a scalable path for Graph Neural Networks (GNNs) in node classification tasks but typically rely on static and rigid routing strategies that enforce a uniform expert budget or coarse-grained expert toggles on all nodes. This limitation overlooks the varying discriminative difficulty of nodes and leads to under-fitting for hard nodes and redundant computation for easy ones. To resolve this issue, we propose D2MoE, a novel framework that shifts the focus from static expert selection to node-wise expert resource allocation. By using predictive entropy as a real-time proxy for difficulty, D2MoE employs a difficulty-driven top-p routing mechanism to adaptively concentrate expert resources on hard nodes while reducing overhead for easy ones, achieving continuous and fine-grained expert budget scaling for node classification. Experiments on 13 benchmarks demonstrate that D2MoE achieves consistent state-of-the-art performance, surpassing leading baselines by up to 7.92% in accuracy on heterophilous graphs. Notably, on large-scale graphs, it reduces memory consumption by up to 73.07% and training time by 46.53% compared to the best-performing Graph MoE, thereby validating its superior efficiency.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。