DeepZipper II: Searching for Lensed Supernovae in Dark Energy Survey Data with Deep Learning

TL;DR

This paper introduces DeepZipper II, a deep learning method for detecting gravitationally lensed supernovae in Dark Energy Survey data, achieving high recall and identifying three candidates for further study.

Contribution

The paper presents a novel multi-branch deep learning architecture trained on simulations to efficiently identify lensed supernovae in large astronomical datasets.

Findings

Achieved 61.13% recall with 0.02% false positive rate on DES data.

Selected 2,245 candidates from over 3 million systems.

Identified three promising lensed supernova candidates for follow-up.

Abstract

Gravitationally lensed supernovae (LSNe) are important probes of cosmic expansion, but they remain rare and difficult to find. Current cosmic surveys likely contain and 5-10 LSNe in total while next-generation experiments are expected to contain several hundreds to a few thousands of these systems. We search for these systems in observed Dark Energy Survey (DES) 5-year SN fields -- 10 3-sq. deg. regions of sky imaged in the $griz$ bands approximately every six nights over five years. To perform the search, we utilize the DeepZipper approach: a multi-branch deep learning architecture trained on image-level simulations of LSNe that simultaneously learns spatial and temporal relationships from time series of images. We find that our method obtains a LSN recall of 61.13% and a false positive rate of 0.02% on the DES SN field data. DeepZipper selected 2,245 candidates from a magnitude-limited ($m_i$ $<$ 22.5) catalog of 3,459,186 systems. We employ human visual inspection to review systems selected by the network and find three candidate LSNe in the DES SN fields.