[论文] FAMOSE: A ReAct Approach to Automated Feature Discovery

论文概要

研究领域: ML
作者: Keith Burghardt, Jienan Liu, Sadman Sakib, Yuning Hao, Bo Li
发布时间: 2026-02-19
arXiv: 2602.17641

中文摘要

特征工程仍是机器学习中的关键瓶颈，尤其对表格数据而言。本文提出FAMOSE（Feature AugMentation and Optimal Selection agEnt），一个基于ReAct范式的新框架，能够自主探索、生成和精炼特征，同时在agent架构中集成特征选择和评估工具。据作者所知，FAMOSE首次将agentic ReAct框架应用于自动化特征工程，同时支持回归和分类任务。大量实验表明，FAMOSE在分类任务上达到或接近SOTA，在回归任务上实现SOTA，平均降低RMSE 2.0%。

原文摘要

Feature engineering remains a critical yet challenging bottleneck in machine learning, particularly for tabular data. We introduce FAMOSE (Feature AugMentation and Optimal Selection agEnt), a novel framework that leverages the ReAct paradigm to autonomously explore, generate, and refine features while integrating feature selection and evaluation tools within an agent architecture. To our knowledge, FAMOSE represents the first application of an agentic ReAct framework to automated feature engineering for both regression and classification tasks. Extensive experiments demonstrate that FAMOSE is at or near the state-of-the-art on classification tasks and achieves SOTA for regression tasks by reducing RMSE by 2.0% on average.

自动采集于 2026-06-24

#论文 #arXiv #ML #小凯