{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:SZDZN7BZROZP3NZJZQ2SI3CZRD","short_pith_number":"pith:SZDZN7BZ","schema_version":"1.0","canonical_sha256":"964796fc398bb2fdb729cc35246c5988cd2e3231174976bf20eb650624609fe3","source":{"kind":"arxiv","id":"2106.00553","version":4},"attestation_state":"computed","paper":{"title":"SHINE: SHaring the INverse Estimate from the forward pass for bi-level optimization and implicit models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Florian Mannel, Jean-Luc Starck, Philippe Ciuciu, Shaojie Bai, Thomas Moreau, Zaccharie Ramzi","submitted_at":"2021-06-01T15:07:34Z","abstract_excerpt":"In recent years, implicit deep learning has emerged as a method to increase the effective depth of deep neural networks. While their training is memory-efficient, they are still significantly slower to train than their explicit counterparts. In Deep Equilibrium Models (DEQs), the training is performed as a bi-level problem, and its computational complexity is partially driven by the iterative inversion of a huge Jacobian matrix. In this paper, we propose a novel strategy to tackle this computational bottleneck from which many bi-level problems suffer. The main idea is to use the quasi-Newton m"},"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":"2106.00553","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-06-01T15:07:34Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"3d980191ceb15c06cd806a5fe1d005122ef3bd6c690e7abe7aeae78699acd4a0","abstract_canon_sha256":"d15311d601bdda79531cce25b48877c223c9cbb94aa732b4b290c1c9ea990fcc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:49:52.376638Z","signature_b64":"1aMmNEM9CU92FCeKyxjDbulltz3xZi3OykhdkId3CSAHymeXWXrs0+tH5/evVwKpsY5Cyu6+9trERnIHXwRaDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"964796fc398bb2fdb729cc35246c5988cd2e3231174976bf20eb650624609fe3","last_reissued_at":"2026-07-05T05:49:52.376236Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:49:52.376236Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SHINE: SHaring the INverse Estimate from the forward pass for bi-level optimization and implicit models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Florian Mannel, Jean-Luc Starck, Philippe Ciuciu, Shaojie Bai, Thomas Moreau, Zaccharie Ramzi","submitted_at":"2021-06-01T15:07:34Z","abstract_excerpt":"In recent years, implicit deep learning has emerged as a method to increase the effective depth of deep neural networks. While their training is memory-efficient, they are still significantly slower to train than their explicit counterparts. In Deep Equilibrium Models (DEQs), the training is performed as a bi-level problem, and its computational complexity is partially driven by the iterative inversion of a huge Jacobian matrix. In this paper, we propose a novel strategy to tackle this computational bottleneck from which many bi-level problems suffer. The main idea is to use the quasi-Newton m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.00553","kind":"arxiv","version":4},"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/2106.00553/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":"2106.00553","created_at":"2026-07-05T05:49:52.376291+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.00553v4","created_at":"2026-07-05T05:49:52.376291+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.00553","created_at":"2026-07-05T05:49:52.376291+00:00"},{"alias_kind":"pith_short_12","alias_value":"SZDZN7BZROZP","created_at":"2026-07-05T05:49:52.376291+00:00"},{"alias_kind":"pith_short_16","alias_value":"SZDZN7BZROZP3NZJ","created_at":"2026-07-05T05:49:52.376291+00:00"},{"alias_kind":"pith_short_8","alias_value":"SZDZN7BZ","created_at":"2026-07-05T05:49:52.376291+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2506.21734","citing_title":"Hierarchical Reasoning Model","ref_index":41,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SZDZN7BZROZP3NZJZQ2SI3CZRD","json":"https://pith.science/pith/SZDZN7BZROZP3NZJZQ2SI3CZRD.json","graph_json":"https://pith.science/api/pith-number/SZDZN7BZROZP3NZJZQ2SI3CZRD/graph.json","events_json":"https://pith.science/api/pith-number/SZDZN7BZROZP3NZJZQ2SI3CZRD/events.json","paper":"https://pith.science/paper/SZDZN7BZ"},"agent_actions":{"view_html":"https://pith.science/pith/SZDZN7BZROZP3NZJZQ2SI3CZRD","download_json":"https://pith.science/pith/SZDZN7BZROZP3NZJZQ2SI3CZRD.json","view_paper":"https://pith.science/paper/SZDZN7BZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.00553&json=true","fetch_graph":"https://pith.science/api/pith-number/SZDZN7BZROZP3NZJZQ2SI3CZRD/graph.json","fetch_events":"https://pith.science/api/pith-number/SZDZN7BZROZP3NZJZQ2SI3CZRD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SZDZN7BZROZP3NZJZQ2SI3CZRD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SZDZN7BZROZP3NZJZQ2SI3CZRD/action/storage_attestation","attest_author":"https://pith.science/pith/SZDZN7BZROZP3NZJZQ2SI3CZRD/action/author_attestation","sign_citation":"https://pith.science/pith/SZDZN7BZROZP3NZJZQ2SI3CZRD/action/citation_signature","submit_replication":"https://pith.science/pith/SZDZN7BZROZP3NZJZQ2SI3CZRD/action/replication_record"}},"created_at":"2026-07-05T05:49:52.376291+00:00","updated_at":"2026-07-05T05:49:52.376291+00:00"}