{"paper":{"title":"Manifold Learning for Source Separation in Confusion-Limited Gravitational-Wave Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Adding manifold normalization to an autoencoder anomaly score improves separation of resolvable sources from LISA's galactic confusion background.","cross_cats":[],"primary_cat":"physics.gen-ph","authors_text":"Jericho Cain","submitted_at":"2025-11-17T00:27:42Z","abstract_excerpt":"The Laser Interferometer Space Antenna (LISA) will observe gravitational waves in a regime that differs sharply from what ground-based detectors such as LIGO handle. Instead of searching for rare signals buried in loud instrumental noise, LISA's main challenge is that its data stream contains millions of unresolved galactic binaries. These blend into a confusion background, and the task becomes identifying sources that stand out from that signal population. We explore whether manifold-learning tools can help with this separation problem. We built a CNN autoencoder trained on the confusion back"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"With this combination, the method achieves an AUC of 0.752, precision 0.81, and recall 0.61, a 35% improvement over the autoencoder alone.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The synthetic LISA data sets used for training and testing, including the modeled confusion background and injected sources, sufficiently resemble the actual data LISA will record so that performance on these simulations predicts performance on flight data.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A CNN autoencoder plus manifold normalization in latent space detects injected sources in synthetic LISA confusion data with AUC 0.752, precision 0.81 and recall 0.61, a 35% gain over autoencoder error alone.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Adding manifold normalization to an autoencoder anomaly score improves separation of resolvable sources from LISA's galactic confusion background.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"4ff2dcfda3b2d7ef106049c74089c8751dc8d2af926e8904dec9111d98b90184"},"source":{"id":"2511.12845","kind":"arxiv","version":4},"verdict":{"id":"22e96703-425e-45bd-9b40-9bf9380124d5","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-17T21:36:53.654340Z","strongest_claim":"With this combination, the method achieves an AUC of 0.752, precision 0.81, and recall 0.61, a 35% improvement over the autoencoder alone.","one_line_summary":"A CNN autoencoder plus manifold normalization in latent space detects injected sources in synthetic LISA confusion data with AUC 0.752, precision 0.81 and recall 0.61, a 35% gain over autoencoder error alone.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The synthetic LISA data sets used for training and testing, including the modeled confusion background and injected sources, sufficiently resemble the actual data LISA will record so that performance on these simulations predicts performance on flight data.","pith_extraction_headline":"Adding manifold normalization to an autoencoder anomaly score improves separation of resolvable sources from LISA's galactic confusion background."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2511.12845/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"6c23972da9b7c1b69c6ddef1be17a59b921b8625a039d9e90576742f6a927563"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}