利用字段依赖关系进行分类数据学习
Exploiting Field Dependencies for Learning on Categorical Data
AI总结：
针对传统分类数据学习未充分利用字段依赖的问题，提出结合全局字段依赖矩阵与样本级局部细化的元学习方法，在六个基准数据集上取得优于现有最优方法的表现。
摘要：
传统的分类数据学习方法未充分利用数据集列（即字段）之间的依赖关系，因为它们仅依赖由分类/回归损失驱动的数据点嵌入。相比之下，我们提出了一种新颖的分类数据学习方法，目标是利用字段之间的依赖关系。我们没有对全局特征统计量进行建模（即通过特征协方差矩阵），而是学习一个捕捉字段间依赖关系的全局字段依赖矩阵，随后在样本层面针对每个字段以不同权重对该全局字段依赖矩阵进行细化（即所谓的局部依赖建模），以改进字段依赖的建模效果。我们的算法采用元学习范式：依赖矩阵在元学习算法的内循环中无需标签即可完成细化，而外循环则以有监督方式（使用标签）同步更新嵌入矩阵（执行投影的矩阵）和全局依赖矩阵。该方法虽简洁，却在六个主流基准数据集上优于多种现有最优方法。详细的消融研究为我们的方法提供了更多洞见。
英文摘要：
Traditional approaches for learning on categorical data underexploit the dependencies between columns (\aka fields) in a dataset because they rely on the embedding of data points driven alone by the classification/regression loss. In contrast, we propose a novel method for learning on categorical data with the goal of exploiting dependencies between fields. Instead of modelling statistics of features globally (i.e., by the covariance matrix of features), we learn a global field dependency matrix that captures dependencies between fields and then we refine the global field dependency matrix at the instance-wise level with different weights (so-called local dependency modelling) w.r.t. each field to improve the modelling of the field dependencies. Our algorithm exploits the meta-learning paradigm, i.e., the dependency matrices are refined in the inner loop of the meta-learning algorithm without the use of labels, whereas the outer loop intertwines the updates of the embedding matrix (the matrix performing projection) and global dependency matrix in a supervised fashion (with the use of labels). Our method is simple yet it outperforms several state-of-the-art methods on six popular dataset benchmarks. Detailed ablation studies provide additional insights into our method.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
