{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:5UNJ4FTMNZDVJNNWNVNXIBY2YL","short_pith_number":"pith:5UNJ4FTM","schema_version":"1.0","canonical_sha256":"ed1a9e166c6e4754b5b66d5b74071ac2de79a30288f75ac1d617884fc4891064","source":{"kind":"arxiv","id":"2203.00089","version":1},"attestation_state":"computed","paper":{"title":"Amortized Proximal Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Jeff Z. HaoChen, Juhan Bae, Paul Vicol, Roger Grosse","submitted_at":"2022-02-28T20:50:48Z","abstract_excerpt":"We propose a framework for online meta-optimization of parameters that govern optimization, called Amortized Proximal Optimization (APO). We first interpret various existing neural network optimizers as approximate stochastic proximal point methods which trade off the current-batch loss with proximity terms in both function space and weight space. The idea behind APO is to amortize the minimization of the proximal point objective by meta-learning the parameters of an update rule. We show how APO can be used to adapt a learning rate or a structured preconditioning matrix. Under appropriate assu"},"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":"2203.00089","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-02-28T20:50:48Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"f94f65cc7c70b92ba6f64c1f0006d833138448c501c83398b7e53e10b0c7316e","abstract_canon_sha256":"3b05a8c561506836a5e2dedcc34c55fb32c96b2f784bbec6e5f4bac97f785a6f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:01:13.631891Z","signature_b64":"qOMEb1fT1WoMSJ7Km6+2thmcXFjnV+Tsf5N1EwbjZ48/V3eoYizTg02URoez4j/nLoTkoRW/EAghnYRnY0EoDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ed1a9e166c6e4754b5b66d5b74071ac2de79a30288f75ac1d617884fc4891064","last_reissued_at":"2026-07-05T04:01:13.631488Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:01:13.631488Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Amortized Proximal Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Jeff Z. HaoChen, Juhan Bae, Paul Vicol, Roger Grosse","submitted_at":"2022-02-28T20:50:48Z","abstract_excerpt":"We propose a framework for online meta-optimization of parameters that govern optimization, called Amortized Proximal Optimization (APO). We first interpret various existing neural network optimizers as approximate stochastic proximal point methods which trade off the current-batch loss with proximity terms in both function space and weight space. The idea behind APO is to amortize the minimization of the proximal point objective by meta-learning the parameters of an update rule. We show how APO can be used to adapt a learning rate or a structured preconditioning matrix. Under appropriate assu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.00089","kind":"arxiv","version":1},"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/2203.00089/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":"2203.00089","created_at":"2026-07-05T04:01:13.631548+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.00089v1","created_at":"2026-07-05T04:01:13.631548+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.00089","created_at":"2026-07-05T04:01:13.631548+00:00"},{"alias_kind":"pith_short_12","alias_value":"5UNJ4FTMNZDV","created_at":"2026-07-05T04:01:13.631548+00:00"},{"alias_kind":"pith_short_16","alias_value":"5UNJ4FTMNZDVJNNW","created_at":"2026-07-05T04:01:13.631548+00:00"},{"alias_kind":"pith_short_8","alias_value":"5UNJ4FTM","created_at":"2026-07-05T04:01:13.631548+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5UNJ4FTMNZDVJNNWNVNXIBY2YL","json":"https://pith.science/pith/5UNJ4FTMNZDVJNNWNVNXIBY2YL.json","graph_json":"https://pith.science/api/pith-number/5UNJ4FTMNZDVJNNWNVNXIBY2YL/graph.json","events_json":"https://pith.science/api/pith-number/5UNJ4FTMNZDVJNNWNVNXIBY2YL/events.json","paper":"https://pith.science/paper/5UNJ4FTM"},"agent_actions":{"view_html":"https://pith.science/pith/5UNJ4FTMNZDVJNNWNVNXIBY2YL","download_json":"https://pith.science/pith/5UNJ4FTMNZDVJNNWNVNXIBY2YL.json","view_paper":"https://pith.science/paper/5UNJ4FTM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.00089&json=true","fetch_graph":"https://pith.science/api/pith-number/5UNJ4FTMNZDVJNNWNVNXIBY2YL/graph.json","fetch_events":"https://pith.science/api/pith-number/5UNJ4FTMNZDVJNNWNVNXIBY2YL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5UNJ4FTMNZDVJNNWNVNXIBY2YL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5UNJ4FTMNZDVJNNWNVNXIBY2YL/action/storage_attestation","attest_author":"https://pith.science/pith/5UNJ4FTMNZDVJNNWNVNXIBY2YL/action/author_attestation","sign_citation":"https://pith.science/pith/5UNJ4FTMNZDVJNNWNVNXIBY2YL/action/citation_signature","submit_replication":"https://pith.science/pith/5UNJ4FTMNZDVJNNWNVNXIBY2YL/action/replication_record"}},"created_at":"2026-07-05T04:01:13.631548+00:00","updated_at":"2026-07-05T04:01:13.631548+00:00"}