Conditional Distribution Estimation Given Functional Covariates Using Deep Operator Networks
Abstract
Functional linear models are commonly used for analyzing functional data with scalar responses, particularly for modeling the conditional mean of the response given functional covariates. In this work, we extend the traditional scalar-on-function regression in two major aspects: 1. estimating the conditional distribution function instead of a particular characteristic such as the conditional mean; 2. considering an arbitrary operator without imposing functional linearity or any model assumption. We use a likelihood approach for the conditional hazard function and apply convolutional neural networks to approximate the effects of functional inputs and estimate the conditional distribution using deep operator networks. Through simulations and a real world data example, we show the desirable robustness of the proposed method in comparison with the mean regression neural networks, demonstrating that our approach achieves better conditional distribution estimation and interval prediction in complex data settings.
Related
- Kernel Density Regression Estimation with Covariates on Riemannian Manifolds
- Outcome-blinded sample size re-estimation for externally controlled single-arm trials using baseline covariates
- Robust Community Detection for Noisy Networks with Covariates: Application to Functional Brain Networks
- Estimation risk in conditional expectiles
- Testing Conditional Distribution Equality Using Generative Models
BibTeX
@misc{hu2026conditional,
title = {{Conditional Distribution Estimation Given Functional Covariates Using Deep Operator Networks}},
author = {Bingqing Hu and Ran Zou and Bin Nan},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.01842},
eprint = {2609.01842},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.01842},
}
RIS
TY - UNPB
TI - Conditional Distribution Estimation Given Functional Covariates Using Deep Operator Networks
AU - Bingqing Hu
AU - Ran Zou
AU - Bin Nan
PY - 2026
DA - 2026/09/03
PB - arXiv
DO - 10.48550/arxiv.2609.01842
UR - https://arxiv.org/abs/2609.01842
AB - Functional linear models are commonly used for analyzing functional data with scalar responses, particularly for modeling the conditional mean of the response given functional covariates. In this work, we extend the traditional scalar-on-function regression in two major aspects: 1. estimating the conditional distribution function instead of a particular characteristic such as the conditional mean; 2. considering an arbitrary operator without imposing functional linearity or any model assumption. We use a likelihood approach for the conditional hazard function and apply convolutional neural networks to approximate the effects of functional inputs and estimate the conditional distribution using deep operator networks. Through simulations and a real world data example, we show the desirable robustness of the proposed method in comparison with the mean regression neural networks, demonstrating that our approach achieves better conditional distribution estimation and interval prediction in complex data settings.
ER -