Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Likelihood-Based One-Class Scoring in CWT Latent Space for Confusion-Limited LISA Gravitational-Wave Detection

Jericho Cain

TL;DR

Results indicate that explicit latent density modeling can outperform local off-manifold distance in this confusion-limited LISA anomaly detection, and provide seed-based comparisons, unified ROC/PR visual summaries, and reproducible experimental artifacts to support follow-on work in LISA anomaly detection.

Abstract

We study one-class scoring for resolvable-source detection in confusion-limited LISA time-series data represented as continuous-wavelet-transform (CWT) scalograms. With data generation and preprocessing held fixed, we benchmark geometry-style scoring against likelihood-style latent-density scoring, while also evaluating morphology-augmented and contrastive variants. Geometry-only and geometry+morphology methods provide modest gains over the reconstruction baseline, and contrastive variants do not show stable improvement. Likelihood scoring on AE latents is consistently stronger: across three seeds, latent-only likelihood reaches ROC-AUC $0.8555\pm 0.0181$ and PR-AUC $0.9219 \pm 0.0118$, versus ROC-AUC $0.7663 \pm 0.0450$ and PR-AUC $0.8667 \pm 0.0255$ for AE+manifold. These results indicate that explicit latent density modeling can outperform local off-manifold distance in this confusion-limited regime. We provide seed-based comparisons, unified ROC/PR visual summaries, and reproducible experimental artifacts to support follow-on work in LISA anomaly detection.

Likelihood-Based One-Class Scoring in CWT Latent Space for Confusion-Limited LISA Gravitational-Wave Detection

TL;DR

Results indicate that explicit latent density modeling can outperform local off-manifold distance in this confusion-limited LISA anomaly detection, and provide seed-based comparisons, unified ROC/PR visual summaries, and reproducible experimental artifacts to support follow-on work in LISA anomaly detection.

Abstract

We study one-class scoring for resolvable-source detection in confusion-limited LISA time-series data represented as continuous-wavelet-transform (CWT) scalograms. With data generation and preprocessing held fixed, we benchmark geometry-style scoring against likelihood-style latent-density scoring, while also evaluating morphology-augmented and contrastive variants. Geometry-only and geometry+morphology methods provide modest gains over the reconstruction baseline, and contrastive variants do not show stable improvement. Likelihood scoring on AE latents is consistently stronger: across three seeds, latent-only likelihood reaches ROC-AUC and PR-AUC , versus ROC-AUC and PR-AUC for AE+manifold. These results indicate that explicit latent density modeling can outperform local off-manifold distance in this confusion-limited regime. We provide seed-based comparisons, unified ROC/PR visual summaries, and reproducible experimental artifacts to support follow-on work in LISA anomaly detection.
Paper Structure (21 sections, 20 equations, 2 figures, 4 tables)

This paper contains 21 sections, 20 equations, 2 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Data and Experimental Protocol
  4. Synthetic benchmark design
  5. Instrumental noise model
  6. Waveform models and parameter draws
  7. Confusion and resolvable-source construction
  8. Dataset composition and reproducibility
  9. Experimental protocol
  10. Methods
  11. Baseline Geometry Framework
  12. Physics-Feature Augmentation
  13. Contrastive Geometry Exploration (Negative Result)
  14. Likelihood One-Class Scoring in Latent Space
  15. Results
  16. ...and 6 more sections

Figures (2)

  • Figure 1: Representative single-run ranking curves across methods. Left: ROC (TPR vs FPR). Right: precision-recall. Legend values correspond to threshold-free summary scores for that run.
  • Figure 2: Latent-likelihood ablation. Left: ROC-AUC and PR-AUC versus GMM component count. Right: ROC-AUC and PR-AUC versus KDE bandwidth (log-scaled). Values are mean $\pm$ std across three seeds.