[2511.04568] Riesz Regression As Direct Density Ratio Estimation

Statistics > Machine Learning arXiv:2511.04568 (stat) [Submitted on 6 Nov 2025 (v1), last revised 24 Mar 2026 (this version, v2)] Title:Riesz Regression As Direct Density Ratio Estimation Authors:Masahiro Kato View a PDF of the paper titled Riesz Regression As Direct Density Ratio Estimation, by Masahiro Kato View PDF HTML (experimental) Abstract:This study clarifies the relationship between Riesz regression [Chernozhukov et al., 2021] and density ratio estimation (DRE) in causal inference problems, such as average treatment effect estimation. We first show that the Riesz representer can be written as a signed density ratio and then demonstrate that the Riesz regression objective coincides with the least-squares importance fitting criterion [Kanamori et al., 2009]. Although Riesz regression applies to a broad class of representer estimation problems, this equivalence with DRE allows us to transfer existing DRE results, including convergence rate analyses, generalizations based on Bregman divergence minimization, and regularization techniques for flexible models such as neural networks. Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG); Econometrics (econ.EM); Statistics Theory (math.ST); Methodology (stat.ME) Cite as: arXiv:2511.04568 [stat.ML]   (or arXiv:2511.04568v2 [stat.ML] for this version)   https://doi.org/10.48550/arXiv.2511.04568 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Masahiro Kato [view email] [v1] Thu, 6 Nov 2025...