用轻量神经代理+离散搜索，提升影响力最大化效率与效果
Rethinking Learning-Based Influence Maximization: Simple Neural Surrogates and Native Discrete Search
- 用统一锚点嵌入和浅层GNN预测感染状态
- 批量多交换模拟退火在无梯度下高效搜索种子集
- 比传统方法更快更准，适合资源受限场景
现有基于学习的影响力最大化方法依赖复杂的神经架构和种子表示的连续优化。我们提出SIMBA，一种与扩散模型无关的框架，结合轻量级神经代理与直接离散搜索。SIMBA引入三个关键组件：1）统一锚点节点嵌入，消除初始化噪声，促使学习聚焦于图拓扑和扩散模式；2）两层浅层图神经网络代理，预测最终感染状态；3）批处理多交换模拟退火算法，在无需梯度或连续松弛的情况下探索组合种子空间。通过将计算重心从复杂表征学习转向高效离散搜索，SIMBA显著缩短求解时间，同时实现更优的影响传播效果和数据效率。代码已开源：https://github.com/yl489/rethink-IM。
原文摘要 · Abstract (English)
Existing learning-based influence maximization frameworks rely heavily on complex neural architectures and continuous optimization over seed representations. We challenge this paradigm with SIMBA, a diffusion-model-agnostic framework pairing a lightweight neural surrogate with direct discrete search. SIMBA introduces three key components: 1) uniformly anchored node embeddings that eliminate initialization noise and encourage learning driven by graph topology and diffusion pattern, 2) a shallow two-layer graph neural network surrogate predicting final infection states, and 3) batched multi-swap simulated annealing that explores combinatorial seed space without gradients or continuous relaxation. By shifting compute from complex representation learning to effective discrete search, SIMBA drastically cuts time-to-solution while achieving superior influence spread and data efficiency. Our code is available at https://github.com/yl489/rethink-IM.
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。