{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:BJ4IBBDJY2YI4LZ27CYZFVWBPB","short_pith_number":"pith:BJ4IBBDJ","schema_version":"1.0","canonical_sha256":"0a78808469c6b08e2f3af8b192d6c17860cb48a06237c99aa33f11f632732de3","source":{"kind":"arxiv","id":"2107.05328","version":2},"attestation_state":"computed","paper":{"title":"Structured Directional Pruning via Perturbation Orthogonal Projection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Qing Wang, Xiaofeng Liu, Yanhui Geng, Yinchuan Li, Yunfeng Shao","submitted_at":"2021-07-12T11:35:47Z","abstract_excerpt":"Structured pruning is an effective compression technique to reduce the computation of neural networks, which is usually achieved by adding perturbations to reduce network parameters at the cost of slightly increasing training loss. A more reasonable approach is to find a sparse minimizer along the flat minimum valley found by optimizers, i.e. stochastic gradient descent, which keeps the training loss constant. To achieve this goal, we propose the structured directional pruning based on orthogonal projecting the perturbations onto the flat minimum valley. We also propose a fast solver sDprun an"},"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":"2107.05328","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-07-12T11:35:47Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"01439596c9a2f0b9dee8cce02afb1d116ca6a467c40f44c9bc662a5402bfdab0","abstract_canon_sha256":"e353abfcab80db8b615f9b9ed952b7dd276f4f4bd54906ee5366ee050432a618"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:24:30.490608Z","signature_b64":"NcKoJZJQ104/RvkmRU5GgmdjVvS61+zZDSYAYxdZqu52OQMslsQjvBf7Vu/Tvo/y4J5rWGlqqFr2azX5AT7KDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0a78808469c6b08e2f3af8b192d6c17860cb48a06237c99aa33f11f632732de3","last_reissued_at":"2026-07-05T03:24:30.490073Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:24:30.490073Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Structured Directional Pruning via Perturbation Orthogonal Projection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Qing Wang, Xiaofeng Liu, Yanhui Geng, Yinchuan Li, Yunfeng Shao","submitted_at":"2021-07-12T11:35:47Z","abstract_excerpt":"Structured pruning is an effective compression technique to reduce the computation of neural networks, which is usually achieved by adding perturbations to reduce network parameters at the cost of slightly increasing training loss. A more reasonable approach is to find a sparse minimizer along the flat minimum valley found by optimizers, i.e. stochastic gradient descent, which keeps the training loss constant. To achieve this goal, we propose the structured directional pruning based on orthogonal projecting the perturbations onto the flat minimum valley. We also propose a fast solver sDprun an"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.05328","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/2107.05328/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":"2107.05328","created_at":"2026-07-05T03:24:30.490137+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.05328v2","created_at":"2026-07-05T03:24:30.490137+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.05328","created_at":"2026-07-05T03:24:30.490137+00:00"},{"alias_kind":"pith_short_12","alias_value":"BJ4IBBDJY2YI","created_at":"2026-07-05T03:24:30.490137+00:00"},{"alias_kind":"pith_short_16","alias_value":"BJ4IBBDJY2YI4LZ2","created_at":"2026-07-05T03:24:30.490137+00:00"},{"alias_kind":"pith_short_8","alias_value":"BJ4IBBDJ","created_at":"2026-07-05T03:24:30.490137+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/BJ4IBBDJY2YI4LZ27CYZFVWBPB","json":"https://pith.science/pith/BJ4IBBDJY2YI4LZ27CYZFVWBPB.json","graph_json":"https://pith.science/api/pith-number/BJ4IBBDJY2YI4LZ27CYZFVWBPB/graph.json","events_json":"https://pith.science/api/pith-number/BJ4IBBDJY2YI4LZ27CYZFVWBPB/events.json","paper":"https://pith.science/paper/BJ4IBBDJ"},"agent_actions":{"view_html":"https://pith.science/pith/BJ4IBBDJY2YI4LZ27CYZFVWBPB","download_json":"https://pith.science/pith/BJ4IBBDJY2YI4LZ27CYZFVWBPB.json","view_paper":"https://pith.science/paper/BJ4IBBDJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.05328&json=true","fetch_graph":"https://pith.science/api/pith-number/BJ4IBBDJY2YI4LZ27CYZFVWBPB/graph.json","fetch_events":"https://pith.science/api/pith-number/BJ4IBBDJY2YI4LZ27CYZFVWBPB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BJ4IBBDJY2YI4LZ27CYZFVWBPB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BJ4IBBDJY2YI4LZ27CYZFVWBPB/action/storage_attestation","attest_author":"https://pith.science/pith/BJ4IBBDJY2YI4LZ27CYZFVWBPB/action/author_attestation","sign_citation":"https://pith.science/pith/BJ4IBBDJY2YI4LZ27CYZFVWBPB/action/citation_signature","submit_replication":"https://pith.science/pith/BJ4IBBDJY2YI4LZ27CYZFVWBPB/action/replication_record"}},"created_at":"2026-07-05T03:24:30.490137+00:00","updated_at":"2026-07-05T03:24:30.490137+00:00"}