PulseAugur
LIVE 08:32:53
research · [2 sources] ·
0
research

Debiased ML: Neyman orthogonal score guides Riesz regression for better balancing

By PulseAugur Editorial · Summary by gemini-2.5-flash-lite from 2 sources

This paper proposes that balancing functions in debiased machine learning should originate from the Neyman orthogonal score, rather than solely relying on covariates. The authors argue that while covariate balancing is suitable when regression error depends only on covariates, it can leave treatment-specific components unbalanced in cases of heterogeneous treatment effects. They advocate for regressor balancing using Riesz regression with basis functions of the full regressor as a more general principle for debiased machine learning. AI

Summary written by gemini-2.5-flash-lite from 2 sources. How we write summaries →

IMPACT Introduces a new theoretical framework for debiased machine learning, potentially improving causal inference in complex datasets.

RANK_REASON This is a research paper published on arXiv discussing a theoretical advancement in debiased machine learning.

Read on arXiv stat.ML →

COVERAGE [2]

  1. arXiv cs.LG TIER_1 · Masahiro Kato ·

    Covariate Balancing and Riesz Regression Should Be Guided by the Neyman Orthogonal Score in Debiased Machine Learning

    arXiv:2605.06386v1 Announce Type: cross Abstract: This position paper argues that, in debiased machine learning, balancing functions should be derived from the Neyman orthogonal score, not chosen only as functions of covariates. Covariate balancing is effective when the regressio…

  2. arXiv stat.ML TIER_1 · Masahiro Kato ·

    Covariate Balancing and Riesz Regression Should Be Guided by the Neyman Orthogonal Score in Debiased Machine Learning

    This position paper argues that, in debiased machine learning, balancing functions should be derived from the Neyman orthogonal score, not chosen only as functions of covariates. Covariate balancing is effective when the regression error entering the score can be represented by f…

RELATED ENTITIES

RELATED TOPICS