Mixture-of-Expert Variational Autoencoders for Cross-Modality Embedding of Type Ia Supernova Data

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shen, Yunyi, Gagliano, Alexander T.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916857292783616
author Shen, Yunyi
Gagliano, Alexander T.
author_facet Shen, Yunyi
Gagliano, Alexander T.
contents Time-domain astrophysics relies on heterogeneous and multi-modal data. Specialized models are often constructed to extract information from a single modality, but this approach ignores the wealth of cross-modality information that may be relevant for the tasks to which the model is applied. In this work, we propose a multi-modal, mixture-of-expert variational autoencoder to learn a joint embedding for supernova light curves and spectra. Our method, which is inspired by the Perceiver architecture, natively accommodates variable-length inputs and the irregular temporal sampling inherent to supernova light curves. We train our model on radiative transfer simulations and validate its performance on cross-modality reconstruction of supernova spectra and physical parameters from the simulation. Our model achieves superior performance in cross-modality generation to nearest-neighbor searches in a contrastively-trained latent space, showing its promise for constructing informative latent representations of multi-modal astronomical datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2507_16817
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Mixture-of-Expert Variational Autoencoders for Cross-Modality Embedding of Type Ia Supernova Data
Shen, Yunyi
Gagliano, Alexander T.
Instrumentation and Methods for Astrophysics
High Energy Astrophysical Phenomena
Time-domain astrophysics relies on heterogeneous and multi-modal data. Specialized models are often constructed to extract information from a single modality, but this approach ignores the wealth of cross-modality information that may be relevant for the tasks to which the model is applied. In this work, we propose a multi-modal, mixture-of-expert variational autoencoder to learn a joint embedding for supernova light curves and spectra. Our method, which is inspired by the Perceiver architecture, natively accommodates variable-length inputs and the irregular temporal sampling inherent to supernova light curves. We train our model on radiative transfer simulations and validate its performance on cross-modality reconstruction of supernova spectra and physical parameters from the simulation. Our model achieves superior performance in cross-modality generation to nearest-neighbor searches in a contrastively-trained latent space, showing its promise for constructing informative latent representations of multi-modal astronomical datasets.
title Mixture-of-Expert Variational Autoencoders for Cross-Modality Embedding of Type Ia Supernova Data
topic Instrumentation and Methods for Astrophysics
High Energy Astrophysical Phenomena
url https://arxiv.org/abs/2507.16817