arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2303.16047cs.LGcs.AIstat.ML
收藏

探索与交互优质稀疏广义加性模型集合

Exploring and Interacting with the Set of Good Sparse Generalized Additive Models

Duke University(杜克大学);The University of British Columbia(不列颠哥伦比亚大学)

Chudi Zhong, Zhi Chen, Jiachang Liu, Margo Seltzer, Cynthia Rudin

AI总结:

本文提出用椭球体近似稀疏广义加性模型的Rashomon集合,以支持用户探索多种近优模型,解决变量重要性、约束建模及形状突变等实际问题。

摘要:

在实际应用中,机器学习模型与领域专家之间的交互至关重要;然而,经典机器学习范式通常只产生单一模型,不利于这种交互。近似并探索Rashomon集合,即所有接近最优模型的集合,通过为用户提供一个可搜索的空间,其中包含多种多样的模型供领域专家选择,从而解决了这一实际挑战。我们提出了算法,用椭球体高效且准确地近似具有固定支持集的稀疏广义加性模型的Rashomon集合,并利用这些椭球体来近似许多不同支持集的Rashomon集合。近似的Rashomon集合作为基石,用于解决实际挑战,例如:(1)研究模型类中变量的重要性;(2)在用户指定约束(单调性、直接编辑)下寻找模型;(3)研究形状函数的突变。实验证明了近似Rashomon集合的保真度及其在解决实际挑战中的有效性。

英文摘要:

In real applications, interaction between machine learning models and domain experts is critical; however, the classical machine learning paradigm that usually produces only a single model does not facilitate such interaction. Approximating and exploring the Rashomon set, i.e., the set of all near-optimal models, addresses this practical challenge by providing the user with a searchable space containing a diverse set of models from which domain experts can choose. We present algorithms to efficiently and accurately approximate the Rashomon set of sparse, generalized additive models with ellipsoids for fixed support sets and use these ellipsoids to approximate Rashomon sets for many different support sets. The approximated Rashomon set serves as a cornerstone to solve practical challenges such as (1) studying the variable importance for the model class; (2) finding models under user-specified constraints (monotonicity, direct editing); and (3) investigating sudden changes in the shape functions. Experiments demonstrate the fidelity of the approximated Rashomon set and its effectiveness in solving practical challenges.

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。

↑