{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:ZTLXZZDQLOQSVKAW4P7HBQ53IA","short_pith_number":"pith:ZTLXZZDQ","schema_version":"1.0","canonical_sha256":"ccd77ce4705ba12aa816e3fe70c3bb4027d3b2da8d66e43b86a379cb2a7ef044","source":{"kind":"arxiv","id":"2211.13692","version":3},"attestation_state":"computed","paper":{"title":"To be or not to be stable, that is the question: understanding neural networks for inverse problems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.NA"],"primary_cat":"math.NA","authors_text":"Davide Evangelista, Elena Loli Piccolomini, Elena Morotti, James Nagy","submitted_at":"2022-11-24T16:16:40Z","abstract_excerpt":"The solution of linear inverse problems arising, for example, in signal and image processing is a challenging problem since the ill-conditioning amplifies, in the solution, the noise present in the data. Recently introduced algorithms based on deep learning overwhelm the more traditional model-based approaches in performance, but they typically suffer from instability with respect to data perturbation. In this paper, we theoretically analyze the trade-off between stability and accuracy of neural networks, when used to solve linear imaging inverse problems for not under-determined cases. Moreov"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2211.13692","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"math.NA","submitted_at":"2022-11-24T16:16:40Z","cross_cats_sorted":["cs.LG","cs.NA"],"title_canon_sha256":"9a0d2f1a7277efaa5cd199e05b1fe5772afaedc4bc7c2a75c23ba9b31bd5c278","abstract_canon_sha256":"0da8e1e00ad9a9e9791b278eb7de4a1a4af93171a1bdc6facb8aae9987c549df"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:42:15.691300Z","signature_b64":"pizyjyT5NoEJzPGPQvqDjdLAU5HpsYAQZPsKHzERfGxmlblt3I0zaiautGoBPkbjWiduH7tad7LmS9Upmo81Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ccd77ce4705ba12aa816e3fe70c3bb4027d3b2da8d66e43b86a379cb2a7ef044","last_reissued_at":"2026-07-05T07:42:15.690858Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:42:15.690858Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"To be or not to be stable, that is the question: understanding neural networks for inverse problems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.NA"],"primary_cat":"math.NA","authors_text":"Davide Evangelista, Elena Loli Piccolomini, Elena Morotti, James Nagy","submitted_at":"2022-11-24T16:16:40Z","abstract_excerpt":"The solution of linear inverse problems arising, for example, in signal and image processing is a challenging problem since the ill-conditioning amplifies, in the solution, the noise present in the data. Recently introduced algorithms based on deep learning overwhelm the more traditional model-based approaches in performance, but they typically suffer from instability with respect to data perturbation. In this paper, we theoretically analyze the trade-off between stability and accuracy of neural networks, when used to solve linear imaging inverse problems for not under-determined cases. Moreov"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.13692","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2211.13692/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":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2211.13692","created_at":"2026-07-05T07:42:15.690913+00:00"},{"alias_kind":"arxiv_version","alias_value":"2211.13692v3","created_at":"2026-07-05T07:42:15.690913+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.13692","created_at":"2026-07-05T07:42:15.690913+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZTLXZZDQLOQS","created_at":"2026-07-05T07:42:15.690913+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZTLXZZDQLOQSVKAW","created_at":"2026-07-05T07:42:15.690913+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZTLXZZDQ","created_at":"2026-07-05T07:42:15.690913+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.09845","citing_title":"Adaptive Weighted Total Variation boosted by learning techniques in few-view tomographic imaging","ref_index":25,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZTLXZZDQLOQSVKAW4P7HBQ53IA","json":"https://pith.science/pith/ZTLXZZDQLOQSVKAW4P7HBQ53IA.json","graph_json":"https://pith.science/api/pith-number/ZTLXZZDQLOQSVKAW4P7HBQ53IA/graph.json","events_json":"https://pith.science/api/pith-number/ZTLXZZDQLOQSVKAW4P7HBQ53IA/events.json","paper":"https://pith.science/paper/ZTLXZZDQ"},"agent_actions":{"view_html":"https://pith.science/pith/ZTLXZZDQLOQSVKAW4P7HBQ53IA","download_json":"https://pith.science/pith/ZTLXZZDQLOQSVKAW4P7HBQ53IA.json","view_paper":"https://pith.science/paper/ZTLXZZDQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2211.13692&json=true","fetch_graph":"https://pith.science/api/pith-number/ZTLXZZDQLOQSVKAW4P7HBQ53IA/graph.json","fetch_events":"https://pith.science/api/pith-number/ZTLXZZDQLOQSVKAW4P7HBQ53IA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZTLXZZDQLOQSVKAW4P7HBQ53IA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZTLXZZDQLOQSVKAW4P7HBQ53IA/action/storage_attestation","attest_author":"https://pith.science/pith/ZTLXZZDQLOQSVKAW4P7HBQ53IA/action/author_attestation","sign_citation":"https://pith.science/pith/ZTLXZZDQLOQSVKAW4P7HBQ53IA/action/citation_signature","submit_replication":"https://pith.science/pith/ZTLXZZDQLOQSVKAW4P7HBQ53IA/action/replication_record"}},"created_at":"2026-07-05T07:42:15.690913+00:00","updated_at":"2026-07-05T07:42:15.690913+00:00"}