Spatio-Spectroscopic Representation Learning using Unsupervised Convolutional Long-Short Term Memory Networks

2026-02-20

AI Summary

This study addresses the challenge of uncovering hidden patterns in galaxy evolution by jointly learning a universal feature representation across spatial and spectral dimensions from integral field spectroscopic data. To this end, we propose an unsupervised deep learning framework based on a Convolutional Long Short-Term Memory (ConvLSTM) autoencoder, applied for the first time to optical emission-line data from approximately 9,000 galaxies in the MaNGA survey. The model enables end-to-end representation learning that seamlessly integrates spatial and spectral information. When evaluated on a subset of 290 active galactic nuclei (AGN), the framework successfully identifies several AGN exhibiting highly anomalous features, thereby demonstrating its effectiveness and potential for unsupervised discovery in astronomical data.

Technology Category

• Machine Learning: Unsupervised & Self-Supervised Learning
• Computer Vision: Representation Learning for Vision
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Integral Field Spectroscopy (IFS) surveys offer a unique new landscape in which to learn in both spatial and spectroscopic dimensions and could help uncover previously unknown insights into galaxy evolution. In this work, we demonstrate a new unsupervised deep learning framework using Convolutional Long-Short Term Memory Network Autoencoders to encode generalized feature representations across both spatial and spectroscopic dimensions spanning $19$ optical emission lines (3800A $<\lambda<$ 8000A) among a sample of $\sim 9000$ galaxies from the MaNGA IFS survey. As a demonstrative exercise, we assess our model on a sample of $290$ Active Galactic Nuclei (AGN) and highlight scientifically interesting characteristics of some highly anomalous AGN.

Problem

• Integral Field Spectroscopy
• Spatio-Spectroscopic Representation
• Galaxy Evolution
• Active Galactic Nuclei
• Unsupervised Learning

Innovation

• Unsupervised Learning
• Convolutional LSTM
• Integral Field Spectroscopy
• Spatio-Spectroscopic Representation
• Galaxy Evolution