Guardado en:
Detalles Bibliográficos
Autores principales: Cheng, Hailing, Yang, Yafang, Tao, Hemeng, Zhang, Fengyu
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:https://arxiv.org/abs/2603.06589
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911624805220352
author Cheng, Hailing
Yang, Yafang
Tao, Hemeng
Zhang, Fengyu
author_facet Cheng, Hailing
Yang, Yafang
Tao, Hemeng
Zhang, Fengyu
contents Model calibration and debiasing are fundamental yet operationally expensive challenges in large-scale recommendation systems. Existing approaches treat them as separate problems requiring distinct infrastructure: post-hoc calibration pipelines, propensity estimation workflows, and per-segment model farms. We introduce the Isotonic Layer, a differentiable piecewise linear module that unifies both problems within a single, lightweight architectural component - requiring no additional data preprocessing, no propensity estimation, and no separate calibration pipelines. The core insight is elegant: by parameterizing non-negative bucket weights as learnable context embeddings, the model automatically learns all calibration and debiasing functions end-to-end from standard training data. Swapping in a different embedding (position, device type, advertiser ID, or any combination) instantly yields calibration tailored to that sub-segment at arbitrary granularity in any high-dimensional feature space, with no engineering changes beyond a single embedding lookup. The same layer handles post-hoc calibration, position debiasing, and heterogeneous multi-task bias correction within one unified framework. This paper offers a principled, practical simplification: a plug-and-play solution that replaces fragmented, high-maintenance calibration infrastructure with a single end-to-end trainable component. Extensive production A/B tests confirm significant improvements in predictive accuracy, calibration fidelity, and ranking consistency.
format Preprint
id arxiv_https___arxiv_org_abs_2603_06589
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Isotonic Layer: A Unified Framework for Recommendation Calibration and Debiasing
Cheng, Hailing
Yang, Yafang
Tao, Hemeng
Zhang, Fengyu
Information Retrieval
Artificial Intelligence
Machine Learning
Model calibration and debiasing are fundamental yet operationally expensive challenges in large-scale recommendation systems. Existing approaches treat them as separate problems requiring distinct infrastructure: post-hoc calibration pipelines, propensity estimation workflows, and per-segment model farms. We introduce the Isotonic Layer, a differentiable piecewise linear module that unifies both problems within a single, lightweight architectural component - requiring no additional data preprocessing, no propensity estimation, and no separate calibration pipelines. The core insight is elegant: by parameterizing non-negative bucket weights as learnable context embeddings, the model automatically learns all calibration and debiasing functions end-to-end from standard training data. Swapping in a different embedding (position, device type, advertiser ID, or any combination) instantly yields calibration tailored to that sub-segment at arbitrary granularity in any high-dimensional feature space, with no engineering changes beyond a single embedding lookup. The same layer handles post-hoc calibration, position debiasing, and heterogeneous multi-task bias correction within one unified framework. This paper offers a principled, practical simplification: a plug-and-play solution that replaces fragmented, high-maintenance calibration infrastructure with a single end-to-end trainable component. Extensive production A/B tests confirm significant improvements in predictive accuracy, calibration fidelity, and ranking consistency.
title Isotonic Layer: A Unified Framework for Recommendation Calibration and Debiasing
topic Information Retrieval
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2603.06589