用低秩适配实现高效神经集成，大幅降低大规模表格数据训练成本
TabLoRA: Parameter-Efficient Low-Rank Ensemble Learning for Large-Scale Tabular Data
- 共享主干网络+低秩适配，避免参数重复
- 在资源受限下性能优于传统树模型和深度学习基线
- 适合需要高效高精度表格建模的工业场景
表格学习仍以梯度提升决策树（GBDT）为主，但近年来深度学习方法日益具备竞争力。然而，在大规模数据集上应用深度表格模型仍具挑战性，因样本量大、特征维度高或类别多会导致显著计算开销。我们提出 TabLoRA，一种面向大规模表格学习的参数高效可训练神经集成方法。不同于完全独立的集成主干，TabLoRA 在预测器间共享通用主干，并引入预测器特异的低秩适配，实现在无完整参数复制的情况下实现集成式预测。在多个基准测试中，相比 GBDT 方法与近期深度学习基线，TabLoRA 在相同资源约束下实现了预测性能与实际效率的良好平衡。内存分析与消融实验进一步表明，该设计提升了神经集成学习的可行性，同时保留了完整集成的大部分优势。
原文摘要 · Abstract (English)
Tabular learning is still dominated by gradient-boosted decision trees (GBDTs), while recent deep learning approaches have become increasingly competitive. However, applying deep tabular models to large-scale datasets remains challenging, as large sample sizes, high feature dimensionality, or many target classes can introduce substantial computational cost. We propose TabLoRA, a parameter-efficient trainable neural ensemble for large-scale tabular learning. Instead of using fully independent ensemble backbones, TabLoRA shares a common backbone across predictors and introduces predictor-specific low-rank adaptations, enabling ensemble-style prediction without full parameter duplication. Across benchmarks, TabLoRA achieves a favorable balance between predictive performance and practical efficiency compared with GBDT methods and recent deep learning baselines under the same resource constraints. Memory analysis and ablation studies further show that the proposed design improves the feasibility of neural ensemble learning while preserving much of the benefit of full ensembles.
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。