BriefGPT.xyz
Nov, 2023
为表格先验数据拟合网络缩放的草图和特征选择
Scaling TabPFN: Sketching and Feature Selection for Tabular Prior-Data Fitted Networks
HTML
PDF
Benjamin Feuer, Chinmay Hegde, Niv Cohen
TL;DR
给定一种预先训练的 TabPFN 对于表格数据,如何将标注的训练样本进行摘要以便输入模型是最佳的方法?我们对 TabPFN 进行了定购和特征选择方法的初步研究,并注意到它与传统拟合表格模型之间的某些关键差异。
Abstract
tabular classification
has traditionally relied on
supervised algorithms
, which estimate the parameters of a prediction model using its training data. Recently,
→