{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:QU4LQL5RGD34UB2HFPR5IKKBAG","short_pith_number":"pith:QU4LQL5R","schema_version":"1.0","canonical_sha256":"8538b82fb130f7ca07472be3d4294101a211ff982516c8966724cb593610819d","source":{"kind":"arxiv","id":"2412.06436","version":3},"attestation_state":"computed","paper":{"title":"An Adaptively Inexact Method for Bilevel Learning Using Primal-Dual Style Differentiation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"math.OC","authors_text":"Hok Shing Wong, Lea Bogensperger, Matthias J. Ehrhardt, Mohammad Sadegh Salehi, Thomas Pock","submitted_at":"2024-12-09T12:26:26Z","abstract_excerpt":"We consider a bilevel learning framework for learning linear operators. In this framework, the learnable parameters are optimized via a loss function that also depends on the minimizer of a convex optimization problem (denoted lower-level problem). We utilize an iterative algorithm called `piggyback' to compute the gradient of the loss and minimizer of the lower-level problem. Given that the lower-level problem is solved numerically, the loss function and thus its gradient can only be computed inexactly. To estimate the accuracy of the computed hypergradient, we derive an a-posteriori error bo"},"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":"2412.06436","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2024-12-09T12:26:26Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"a61a9fa86d28484b1752e6439aa823266040113af3a3978046cc126ca372f93f","abstract_canon_sha256":"716f88f1bc197d9b7558103019fe53091d161241695aaf167f06451eb8572521"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:17:38.714617Z","signature_b64":"GmJAoJHyQPH9sBxvR91fn5K75fOFBs9u68GUSQxu1jGi5cw9q+cvo8IMauuFVMScEvxn/L4YtAHsirsdAp4UAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8538b82fb130f7ca07472be3d4294101a211ff982516c8966724cb593610819d","last_reissued_at":"2026-07-05T11:17:38.714045Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:17:38.714045Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"An Adaptively Inexact Method for Bilevel Learning Using Primal-Dual Style Differentiation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"math.OC","authors_text":"Hok Shing Wong, Lea Bogensperger, Matthias J. Ehrhardt, Mohammad Sadegh Salehi, Thomas Pock","submitted_at":"2024-12-09T12:26:26Z","abstract_excerpt":"We consider a bilevel learning framework for learning linear operators. In this framework, the learnable parameters are optimized via a loss function that also depends on the minimizer of a convex optimization problem (denoted lower-level problem). We utilize an iterative algorithm called `piggyback' to compute the gradient of the loss and minimizer of the lower-level problem. Given that the lower-level problem is solved numerically, the loss function and thus its gradient can only be computed inexactly. To estimate the accuracy of the computed hypergradient, we derive an a-posteriori error bo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.06436","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/2412.06436/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":"2412.06436","created_at":"2026-07-05T11:17:38.714113+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.06436v3","created_at":"2026-07-05T11:17:38.714113+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.06436","created_at":"2026-07-05T11:17:38.714113+00:00"},{"alias_kind":"pith_short_12","alias_value":"QU4LQL5RGD34","created_at":"2026-07-05T11:17:38.714113+00:00"},{"alias_kind":"pith_short_16","alias_value":"QU4LQL5RGD34UB2H","created_at":"2026-07-05T11:17:38.714113+00:00"},{"alias_kind":"pith_short_8","alias_value":"QU4LQL5R","created_at":"2026-07-05T11:17:38.714113+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.13239","citing_title":"Restarted contractive operators to learn at equilibrium","ref_index":6,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QU4LQL5RGD34UB2HFPR5IKKBAG","json":"https://pith.science/pith/QU4LQL5RGD34UB2HFPR5IKKBAG.json","graph_json":"https://pith.science/api/pith-number/QU4LQL5RGD34UB2HFPR5IKKBAG/graph.json","events_json":"https://pith.science/api/pith-number/QU4LQL5RGD34UB2HFPR5IKKBAG/events.json","paper":"https://pith.science/paper/QU4LQL5R"},"agent_actions":{"view_html":"https://pith.science/pith/QU4LQL5RGD34UB2HFPR5IKKBAG","download_json":"https://pith.science/pith/QU4LQL5RGD34UB2HFPR5IKKBAG.json","view_paper":"https://pith.science/paper/QU4LQL5R","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.06436&json=true","fetch_graph":"https://pith.science/api/pith-number/QU4LQL5RGD34UB2HFPR5IKKBAG/graph.json","fetch_events":"https://pith.science/api/pith-number/QU4LQL5RGD34UB2HFPR5IKKBAG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QU4LQL5RGD34UB2HFPR5IKKBAG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QU4LQL5RGD34UB2HFPR5IKKBAG/action/storage_attestation","attest_author":"https://pith.science/pith/QU4LQL5RGD34UB2HFPR5IKKBAG/action/author_attestation","sign_citation":"https://pith.science/pith/QU4LQL5RGD34UB2HFPR5IKKBAG/action/citation_signature","submit_replication":"https://pith.science/pith/QU4LQL5RGD34UB2HFPR5IKKBAG/action/replication_record"}},"created_at":"2026-07-05T11:17:38.714113+00:00","updated_at":"2026-07-05T11:17:38.714113+00:00"}