{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:JTKM6TOGZQNNGLYPWEPHLNXCKO","short_pith_number":"pith:JTKM6TOG","schema_version":"1.0","canonical_sha256":"4cd4cf4dc6cc1ad32f0fb11e75b6e25395ed614b862005a124f1ee55eb616957","source":{"kind":"arxiv","id":"2502.02496","version":2},"attestation_state":"computed","paper":{"title":"Deep Weight Factorization: Sparse Learning Through the Lens of Artificial Symmetries","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Bernd Bischl, Chris Kolb, David R\\\"ugamer, Tobias Weber","submitted_at":"2025-02-04T17:12:56Z","abstract_excerpt":"Sparse regularization techniques are well-established in machine learning, yet their application in neural networks remains challenging due to the non-differentiability of penalties like the $L_1$ norm, which is incompatible with stochastic gradient descent. A promising alternative is shallow weight factorization, where weights are decomposed into two factors, allowing for smooth optimization of $L_1$-penalized neural networks by adding differentiable $L_2$ regularization to the factors. In this work, we introduce deep weight factorization, extending previous shallow approaches to more than tw"},"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":"2502.02496","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-04T17:12:56Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"db08034f869d2b559b295977951443c5ec954095d2f0424ef151386b3f258386","abstract_canon_sha256":"0aa61212864fcd0bec293f9a082c72425be5f5e74433202c1386b9cafea8de7b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:10:49.180829Z","signature_b64":"6s5dbunrTpS1vJA9q7O3SdtD8cdOo3sfIBn8hKDNQB2KG07bTAkKc5YVYBY6RtZhva6Xz1XtC6XUiYdAVOjUBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4cd4cf4dc6cc1ad32f0fb11e75b6e25395ed614b862005a124f1ee55eb616957","last_reissued_at":"2026-07-05T10:10:49.180365Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:10:49.180365Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Deep Weight Factorization: Sparse Learning Through the Lens of Artificial Symmetries","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Bernd Bischl, Chris Kolb, David R\\\"ugamer, Tobias Weber","submitted_at":"2025-02-04T17:12:56Z","abstract_excerpt":"Sparse regularization techniques are well-established in machine learning, yet their application in neural networks remains challenging due to the non-differentiability of penalties like the $L_1$ norm, which is incompatible with stochastic gradient descent. A promising alternative is shallow weight factorization, where weights are decomposed into two factors, allowing for smooth optimization of $L_1$-penalized neural networks by adding differentiable $L_2$ regularization to the factors. In this work, we introduce deep weight factorization, extending previous shallow approaches to more than tw"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.02496","kind":"arxiv","version":2},"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/2502.02496/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":"2502.02496","created_at":"2026-07-05T10:10:49.180423+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.02496v2","created_at":"2026-07-05T10:10:49.180423+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.02496","created_at":"2026-07-05T10:10:49.180423+00:00"},{"alias_kind":"pith_short_12","alias_value":"JTKM6TOGZQNN","created_at":"2026-07-05T10:10:49.180423+00:00"},{"alias_kind":"pith_short_16","alias_value":"JTKM6TOGZQNNGLYP","created_at":"2026-07-05T10:10:49.180423+00:00"},{"alias_kind":"pith_short_8","alias_value":"JTKM6TOG","created_at":"2026-07-05T10:10:49.180423+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.03547","citing_title":"Erase Persona, Forget Lore: Benchmarking Multimodal Copyright Unlearning in Large Vision Language Models","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18152","citing_title":"mlr3torch: A Deep Learning Framework in R based on mlr3 and torch","ref_index":45,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JTKM6TOGZQNNGLYPWEPHLNXCKO","json":"https://pith.science/pith/JTKM6TOGZQNNGLYPWEPHLNXCKO.json","graph_json":"https://pith.science/api/pith-number/JTKM6TOGZQNNGLYPWEPHLNXCKO/graph.json","events_json":"https://pith.science/api/pith-number/JTKM6TOGZQNNGLYPWEPHLNXCKO/events.json","paper":"https://pith.science/paper/JTKM6TOG"},"agent_actions":{"view_html":"https://pith.science/pith/JTKM6TOGZQNNGLYPWEPHLNXCKO","download_json":"https://pith.science/pith/JTKM6TOGZQNNGLYPWEPHLNXCKO.json","view_paper":"https://pith.science/paper/JTKM6TOG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.02496&json=true","fetch_graph":"https://pith.science/api/pith-number/JTKM6TOGZQNNGLYPWEPHLNXCKO/graph.json","fetch_events":"https://pith.science/api/pith-number/JTKM6TOGZQNNGLYPWEPHLNXCKO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JTKM6TOGZQNNGLYPWEPHLNXCKO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JTKM6TOGZQNNGLYPWEPHLNXCKO/action/storage_attestation","attest_author":"https://pith.science/pith/JTKM6TOGZQNNGLYPWEPHLNXCKO/action/author_attestation","sign_citation":"https://pith.science/pith/JTKM6TOGZQNNGLYPWEPHLNXCKO/action/citation_signature","submit_replication":"https://pith.science/pith/JTKM6TOGZQNNGLYPWEPHLNXCKO/action/replication_record"}},"created_at":"2026-07-05T10:10:49.180423+00:00","updated_at":"2026-07-05T10:10:49.180423+00:00"}