{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:JSDZTHYHVFEUCFIT4ZVYEGD25W","short_pith_number":"pith:JSDZTHYH","schema_version":"1.0","canonical_sha256":"4c87999f07a949411513e66b82187aedb00792dff57407743474d7a7dd1087b6","source":{"kind":"arxiv","id":"2005.03354","version":2},"attestation_state":"computed","paper":{"title":"DMCP: Differentiable Markov Channel Pruning for Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Junjie Yan, Quanquan Li, Shaopeng Guo, Yujie Wang","submitted_at":"2020-05-07T09:39:55Z","abstract_excerpt":"Recent works imply that the channel pruning can be regarded as searching optimal sub-structure from unpruned networks. However, existing works based on this observation require training and evaluating a large number of structures, which limits their application. In this paper, we propose a novel differentiable method for channel pruning, named Differentiable Markov Channel Pruning (DMCP), to efficiently search the optimal sub-structure. Our method is differentiable and can be directly optimized by gradient descent with respect to standard task loss and budget regularization (e.g. FLOPs constra"},"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":"2005.03354","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2020-05-07T09:39:55Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"16ec569b0651682443164992a99985b23804d6f1625db250516e9ae7e35ba90f","abstract_canon_sha256":"a397edc8a34bc1c9e732e184b55f93a2848c5760ae9b394faa733c1e223b87f6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:01:24.994671Z","signature_b64":"0CQ/dTFs8Sl3oOA+vRo0I8btfxD0dhO/p+7Zkml1r+1rsv83tRRkVobJdnaTbB3z50VwUH7eq8QJ9T/abaF8Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4c87999f07a949411513e66b82187aedb00792dff57407743474d7a7dd1087b6","last_reissued_at":"2026-07-05T01:01:24.994257Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:01:24.994257Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DMCP: Differentiable Markov Channel Pruning for Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Junjie Yan, Quanquan Li, Shaopeng Guo, Yujie Wang","submitted_at":"2020-05-07T09:39:55Z","abstract_excerpt":"Recent works imply that the channel pruning can be regarded as searching optimal sub-structure from unpruned networks. However, existing works based on this observation require training and evaluating a large number of structures, which limits their application. In this paper, we propose a novel differentiable method for channel pruning, named Differentiable Markov Channel Pruning (DMCP), to efficiently search the optimal sub-structure. Our method is differentiable and can be directly optimized by gradient descent with respect to standard task loss and budget regularization (e.g. FLOPs constra"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2005.03354","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/2005.03354/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":"2005.03354","created_at":"2026-07-05T01:01:24.994336+00:00"},{"alias_kind":"arxiv_version","alias_value":"2005.03354v2","created_at":"2026-07-05T01:01:24.994336+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2005.03354","created_at":"2026-07-05T01:01:24.994336+00:00"},{"alias_kind":"pith_short_12","alias_value":"JSDZTHYHVFEU","created_at":"2026-07-05T01:01:24.994336+00:00"},{"alias_kind":"pith_short_16","alias_value":"JSDZTHYHVFEUCFIT","created_at":"2026-07-05T01:01:24.994336+00:00"},{"alias_kind":"pith_short_8","alias_value":"JSDZTHYH","created_at":"2026-07-05T01:01:24.994336+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/JSDZTHYHVFEUCFIT4ZVYEGD25W","json":"https://pith.science/pith/JSDZTHYHVFEUCFIT4ZVYEGD25W.json","graph_json":"https://pith.science/api/pith-number/JSDZTHYHVFEUCFIT4ZVYEGD25W/graph.json","events_json":"https://pith.science/api/pith-number/JSDZTHYHVFEUCFIT4ZVYEGD25W/events.json","paper":"https://pith.science/paper/JSDZTHYH"},"agent_actions":{"view_html":"https://pith.science/pith/JSDZTHYHVFEUCFIT4ZVYEGD25W","download_json":"https://pith.science/pith/JSDZTHYHVFEUCFIT4ZVYEGD25W.json","view_paper":"https://pith.science/paper/JSDZTHYH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2005.03354&json=true","fetch_graph":"https://pith.science/api/pith-number/JSDZTHYHVFEUCFIT4ZVYEGD25W/graph.json","fetch_events":"https://pith.science/api/pith-number/JSDZTHYHVFEUCFIT4ZVYEGD25W/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JSDZTHYHVFEUCFIT4ZVYEGD25W/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JSDZTHYHVFEUCFIT4ZVYEGD25W/action/storage_attestation","attest_author":"https://pith.science/pith/JSDZTHYHVFEUCFIT4ZVYEGD25W/action/author_attestation","sign_citation":"https://pith.science/pith/JSDZTHYHVFEUCFIT4ZVYEGD25W/action/citation_signature","submit_replication":"https://pith.science/pith/JSDZTHYHVFEUCFIT4ZVYEGD25W/action/replication_record"}},"created_at":"2026-07-05T01:01:24.994336+00:00","updated_at":"2026-07-05T01:01:24.994336+00:00"}