Removing Hidden Confounding by Experimental Grounding

2018-10-27 · NeurIPS 2018 12 · Nathan Kallus, Aahlad Manas Puli, Uri Shalit

Observational data is increasingly used as a means for making individual-level causal predictions and intervention recommendations. The foremost challenge of causal inference from observational data is hidden confounding, whose presence cannot be tested in data and can invalidate any causal conclusion. Experimental data does not suffer from confounding but is usually limited in both scope and scale. We introduce a novel method of using limited experimental data to correct the hidden confounding in causal effect models trained on larger observational data, even if the observational data does not fully overlap with the experimental data. Our method makes strictly weaker assumptions than existing approaches, and we prove conditions under which it yields a consistent estimator. We demonstrate our method's efficacy using real-world data from a large educational experiment.

📄 PDF Abstract BibTeX arXiv:1810.11646

Code (0)

등록된 구현이 없습니다.

Tasks

Causal Inference

Methods 이 논문이 사용한 방법론

Causal inference Causal inference is the process of drawing a conclusion about a causal connection based on the conditions of the occurrence of an effect. The main difference between causal…

Removing Hidden Confounding in Recommendation: A Unified Multi-Task Learning Approach

2023-09-21 · NeurIPS 2023 11

In recommender systems, the collected data used for training is always subject to selection bias, which poses a great challenge for unbiased learning. Previous studies proposed various debiasing methods based on observed…

Deconfounding and Causal Regularization for Stability and External Validity

2020-08-14 · Peter Bühlmann, Domagoj Ćevid

We review some recent work on removing hidden confounding and causal regularization from a unified viewpoint. We describe how simple and user-friendly techniques improve stability, replicability and distributional robust…

Delphic Offline Reinforcement Learning under Nonidentifiable Hidden Confounding

2023-06-01 · Alizée Pace, Hugo Yèche, Bernhard Schölkopf, Gunnar Rätsch 외

A prominent challenge of offline reinforcement learning (RL) is the issue of hidden confounding: unobserved variables may influence both the actions taken by the agent and the observed outcomes. Hidden confounding can co…

Management Offline RL reinforcement-learning Reinforcement Learning+1

When Prompts Interact: Assessing Prompt Arithmetic for Deconfounding under Distribution Shift

2026-05-04 · Zhecheng Sheng, Yongsen Tan, Xiruo Ding, Trevor Cohen 외 arxiv

In classification tasks, models may rely on confounding variables to achieve strong in-distribution performance, capturing spurious features that fail under distribution shift. This shortcut behavior leads to substantial…

Robust Personalized Recommendation under Hidden Confounding in MNAR

2026-05-20 · Zongyu Li, Wanting Su, Tianyu Xia arxiv

Recommender systems often rely on observational user--item interaction data, which is prone to selection bias due to users' selective interactions with items. Inverse propensity weighting and doubly robust estimators eff…