Zephyr : Stitching Heterogeneous Training Data with Normalizing Flows for Photometric Redshift Inference

TL;DR

Zephyr introduces a novel normalizing flow-based framework for effectively integrating heterogeneous training data in photometric redshift inference, improving robustness, interpretability, and uncertainty quantification.

Contribution

It is the first to combine normalizing flows with mixture density estimation for heterogeneous data in photometric redshift inference, enhancing robustness and interpretability.

Findings

Improved robustness in point and distribution estimates.

Explicit disentanglement of multi-source data contributions.

Enhanced uncertainty quantification and interpretability.

Abstract

We present zephyr, a novel method that integrates cutting-edge normalizing flow techniques into a mixture density estimation framework, enabling the effective use of heterogeneous training data for photometric redshift inference. Compared to previous methods, zephyr demonstrates enhanced robustness for both point estimation and distribution reconstruction by leveraging normalizing flows for density estimation and incorporating careful uncertainty quantification. Moreover, zephyr offers unique interpretability by explicitly disentangling contributions from multi-source training data, which can facilitate future weak lensing analysis by providing an additional quality assessment. As probabilistic generative deep learning techniques gain increasing prominence in astronomy, zephyr should become an inspiration for handling heterogeneous training data while remaining interpretable and robustly accounting for observational uncertainties.