{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:JASCWLRSPZHDMUGVFZIG7HRQFF","short_pith_number":"pith:JASCWLRS","schema_version":"1.0","canonical_sha256":"48242b2e327e4e3650d52e506f9e30297fc267975e73d673a91115ff376ca82b","source":{"kind":"arxiv","id":"2202.08132","version":2},"attestation_state":"computed","paper":{"title":"Prospect Pruning: Finding Trainable Weights at Initialization using Meta-Gradients","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Joost van Amersfoort, Luisa M Zintgraf, Milad Alizadeh, Nicholas Donald Lane, Sebastian Farquhar, Shyam A. Tailor, Yarin Gal","submitted_at":"2022-02-16T15:18:55Z","abstract_excerpt":"Pruning neural networks at initialization would enable us to find sparse models that retain the accuracy of the original network while consuming fewer computational resources for training and inference. However, current methods are insufficient to enable this optimization and lead to a large degradation in model performance. In this paper, we identify a fundamental limitation in the formulation of current methods, namely that their saliency criteria look at a single step at the start of training without taking into account the trainability of the network. While pruning iteratively and graduall"},"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":"2202.08132","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-02-16T15:18:55Z","cross_cats_sorted":[],"title_canon_sha256":"32f8c4fb135c393c5df096542f5a69443d68cfe09d3776d846ec1fd3bfbe5754","abstract_canon_sha256":"3e6ce9bdabfbecb5ed1259dc4bcda29744a11c6f240f5e4ca790330f93602b6e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:11:30.495127Z","signature_b64":"PukQjiwLlQouulbVt8Pm/kziblRHVXhXZMULtNsH02fOmjwn5f61AAbOKWR14e+Mc8MM5KyytcMMMBphBnZyCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"48242b2e327e4e3650d52e506f9e30297fc267975e73d673a91115ff376ca82b","last_reissued_at":"2026-07-05T04:11:30.494781Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:11:30.494781Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Prospect Pruning: Finding Trainable Weights at Initialization using Meta-Gradients","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Joost van Amersfoort, Luisa M Zintgraf, Milad Alizadeh, Nicholas Donald Lane, Sebastian Farquhar, Shyam A. Tailor, Yarin Gal","submitted_at":"2022-02-16T15:18:55Z","abstract_excerpt":"Pruning neural networks at initialization would enable us to find sparse models that retain the accuracy of the original network while consuming fewer computational resources for training and inference. However, current methods are insufficient to enable this optimization and lead to a large degradation in model performance. In this paper, we identify a fundamental limitation in the formulation of current methods, namely that their saliency criteria look at a single step at the start of training without taking into account the trainability of the network. While pruning iteratively and graduall"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2202.08132","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/2202.08132/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":"2202.08132","created_at":"2026-07-05T04:11:30.494832+00:00"},{"alias_kind":"arxiv_version","alias_value":"2202.08132v2","created_at":"2026-07-05T04:11:30.494832+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2202.08132","created_at":"2026-07-05T04:11:30.494832+00:00"},{"alias_kind":"pith_short_12","alias_value":"JASCWLRSPZHD","created_at":"2026-07-05T04:11:30.494832+00:00"},{"alias_kind":"pith_short_16","alias_value":"JASCWLRSPZHDMUGV","created_at":"2026-07-05T04:11:30.494832+00:00"},{"alias_kind":"pith_short_8","alias_value":"JASCWLRS","created_at":"2026-07-05T04:11:30.494832+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/JASCWLRSPZHDMUGVFZIG7HRQFF","json":"https://pith.science/pith/JASCWLRSPZHDMUGVFZIG7HRQFF.json","graph_json":"https://pith.science/api/pith-number/JASCWLRSPZHDMUGVFZIG7HRQFF/graph.json","events_json":"https://pith.science/api/pith-number/JASCWLRSPZHDMUGVFZIG7HRQFF/events.json","paper":"https://pith.science/paper/JASCWLRS"},"agent_actions":{"view_html":"https://pith.science/pith/JASCWLRSPZHDMUGVFZIG7HRQFF","download_json":"https://pith.science/pith/JASCWLRSPZHDMUGVFZIG7HRQFF.json","view_paper":"https://pith.science/paper/JASCWLRS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2202.08132&json=true","fetch_graph":"https://pith.science/api/pith-number/JASCWLRSPZHDMUGVFZIG7HRQFF/graph.json","fetch_events":"https://pith.science/api/pith-number/JASCWLRSPZHDMUGVFZIG7HRQFF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JASCWLRSPZHDMUGVFZIG7HRQFF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JASCWLRSPZHDMUGVFZIG7HRQFF/action/storage_attestation","attest_author":"https://pith.science/pith/JASCWLRSPZHDMUGVFZIG7HRQFF/action/author_attestation","sign_citation":"https://pith.science/pith/JASCWLRSPZHDMUGVFZIG7HRQFF/action/citation_signature","submit_replication":"https://pith.science/pith/JASCWLRSPZHDMUGVFZIG7HRQFF/action/replication_record"}},"created_at":"2026-07-05T04:11:30.494832+00:00","updated_at":"2026-07-05T04:11:30.494832+00:00"}