论文标题
DNF-NET：表格数据的神经体系结构
DNF-Net: A Neural Architecture for Tabular Data
论文作者
论文摘要
深度学习中一个充满挑战的开放问题是如何处理表格数据。与诸如图像和自然语言处理（深度体系结构占上风）之类的领域不同，仍然没有广泛接受的神经体系结构来主导表格数据。作为弥合这一差距的一步，我们提出了DNF-NET一种新颖的通用结构，其感应性偏置模型与仿射软阈值决策术语相对于脱节的正常形式（DNF），其结构对应于逻辑布尔公式。此外，DNF-NET促进了局部决策，这些决策是在特征的小子集上做出的。我们提出了一项广泛的经验研究，表明DNF-NET与表格数据相比，表现明显优于FCN。对于相对较少的超参数，DNF-NETS使用神经网络为对表格数据的实际端到端处理打开了大门。我们提出了消融研究，该研究证明了DNF-NET的设计选择合理，包括三个电感偏置元素，即布尔公式，位置和特征选择。
A challenging open question in deep learning is how to handle tabular data. Unlike domains such as image and natural language processing, where deep architectures prevail, there is still no widely accepted neural architecture that dominates tabular data. As a step toward bridging this gap, we present DNF-Net a novel generic architecture whose inductive bias elicits models whose structure corresponds to logical Boolean formulas in disjunctive normal form (DNF) over affine soft-threshold decision terms. In addition, DNF-Net promotes localized decisions that are taken over small subsets of the features. We present an extensive empirical study showing that DNF-Nets significantly and consistently outperform FCNs over tabular data. With relatively few hyperparameters, DNF-Nets open the door to practical end-to-end handling of tabular data using neural networks. We present ablation studies, which justify the design choices of DNF-Net including the three inductive bias elements, namely, Boolean formulation, locality, and feature selection.
