用粒度球提升随机向量网络的鲁棒性与可扩展性
GB-RVFL: Fusion of Randomized Neural Network and Granular Ball Computing
- 用粒度球替代原始样本，降低计算复杂度
- 仅需逆运算球心矩阵，提升大规模数据处理能力
- 融合图嵌入保留数据拓扑结构，适合高噪声场景
随机向量函数链接（RVFL）网络具有强大的泛化能力，但对所有样本一视同仁，忽略噪声影响，且因需逆运算整个训练矩阵而扩展性受限。为此，本文提出粒度球RVFL（GB-RVFL）模型，以粒度球（GBs）作为输入，仅需逆运算球心矩阵，显著提升可扩展性，并通过粒度粗粒度增强抗噪与抗异常值能力。此外，为捕捉数据几何结构，进一步提出图嵌入GB-RVFL（GE-GB-RVFL）模型，融合粒度计算与图嵌入（GE），保持粒度球间的拓扑关系。在KEEL、UCI、NDC及生物医学数据集上的实验表明，所提模型性能优于基线方法。
原文摘要 · Abstract (English)
The random vector functional link (RVFL) network is a prominent classification model with strong generalization ability. However, RVFL treats all samples uniformly, ignoring whether they are pure or noisy, and its scalability is limited due to the need for inverting the entire training matrix. To address these issues, we propose granular ball RVFL (GB-RVFL) model, which uses granular balls (GBs) as inputs instead of training samples. This approach enhances scalability by requiring only the inverse of the GB center matrix and improves robustness against noise and outliers through the coarse granularity of GBs. Furthermore, RVFL overlooks the dataset's geometric structure. To address this, we propose graph embedding GB-RVFL (GE-GB-RVFL) model, which fuses granular computing and graph embedding (GE) to preserve the topological structure of GBs. The proposed GB-RVFL and GE-GB-RVFL models are evaluated on KEEL, UCI, NDC and biomedical datasets, demonstrating superior performance compared to baseline models.
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。