A contrastive self-supervised convolutional autoencoder detects core-collapse supernova gravitational waves with performance comparable to supervised CNNs, better generalization to unseen waveforms, and ~120 kpc sensitive distance under Einstein Telescope noise.
Rotating stellar core-collapse waveform decompositon: a Principal Component Analysis approach
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
This paper introduces the use of Principal Component Analysis as a method to decompose the waveform catalogues to produce a set of orthonormal basis vectors. We apply this method to a set of supernova waveforms and compare the basis vectors obtained with those obtained through Gram-Schmidt decomposition. We observe that, for the chosen set of waveforms, the performance of the two methods are comparable for minimal match requirements up to 0.9, with 14 Gram-Schmidt basis vectors and 12 principal components required for a minimal match of 0.9. This implies that there are many common features in the chosen waveforms. Additionally, we observe the chosen waveforms have very similar features and a minimal match of 0.7 can be obtained by decomposing only one third of the entire set of waveforms in the chosen catalogue. We discuss the implications of this observation and the advantages of eigen-decomposing waveform catalogues with Principal Component Analysis.
citation-role summary
background 1
citation-polarity summary
fields
gr-qc 1years
2026 1verdicts
UNVERDICTED 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Contrastive self-supervised convolutional autoencoder for core-collapse supernova gravitational-wave detection
A contrastive self-supervised convolutional autoencoder detects core-collapse supernova gravitational waves with performance comparable to supervised CNNs, better generalization to unseen waveforms, and ~120 kpc sensitive distance under Einstein Telescope noise.