{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:5QYSIGSCYI7RAB46PKENETTT65","short_pith_number":"pith:5QYSIGSC","schema_version":"1.0","canonical_sha256":"ec31241a42c23f10079e7a88d24e73f75c7ea28353c20a20cc1cca795dafd9b8","source":{"kind":"arxiv","id":"1705.07356","version":4},"attestation_state":"computed","paper":{"title":"Structural Compression of Convolutional Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bin Yu, Reza Abbasi-Asl","submitted_at":"2017-05-20T20:12:07Z","abstract_excerpt":"Deep convolutional neural networks (CNNs) have been successful in many tasks in machine vision, however, millions of weights in the form of thousands of convolutional filters in CNNs makes them difficult for human intepretation or understanding in science. In this article, we introduce CAR, a greedy structural compression scheme to obtain smaller and more interpretable CNNs, while achieving close to original accuracy. The compression is based on pruning filters with the least contribution to the classification accuracy. We demonstrate the interpretability of CAR-compressed CNNs by showing that"},"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":"1705.07356","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2017-05-20T20:12:07Z","cross_cats_sorted":[],"title_canon_sha256":"cfeab63094883da822d555a504b16fbfd28d40610dc9d93323f49f1cd30ac74a","abstract_canon_sha256":"68aca3c1b6b75c52887924ecbb50dba59b4f6e95fa0500809768a58db834860a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:50:23.163032Z","signature_b64":"Nx2ytVU0F+0zEvyMtpq6dE7cLD4ZD8FeDfdZTJ+tKBee5aYXDPvE+moIyUkfu4SsdBc0nt7cmYA1iYupSznaBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ec31241a42c23f10079e7a88d24e73f75c7ea28353c20a20cc1cca795dafd9b8","last_reissued_at":"2026-07-05T00:50:23.162559Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:50:23.162559Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Structural Compression of Convolutional Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bin Yu, Reza Abbasi-Asl","submitted_at":"2017-05-20T20:12:07Z","abstract_excerpt":"Deep convolutional neural networks (CNNs) have been successful in many tasks in machine vision, however, millions of weights in the form of thousands of convolutional filters in CNNs makes them difficult for human intepretation or understanding in science. In this article, we introduce CAR, a greedy structural compression scheme to obtain smaller and more interpretable CNNs, while achieving close to original accuracy. The compression is based on pruning filters with the least contribution to the classification accuracy. We demonstrate the interpretability of CAR-compressed CNNs by showing that"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1705.07356","kind":"arxiv","version":4},"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/1705.07356/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":"1705.07356","created_at":"2026-07-05T00:50:23.162624+00:00"},{"alias_kind":"arxiv_version","alias_value":"1705.07356v4","created_at":"2026-07-05T00:50:23.162624+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1705.07356","created_at":"2026-07-05T00:50:23.162624+00:00"},{"alias_kind":"pith_short_12","alias_value":"5QYSIGSCYI7R","created_at":"2026-07-05T00:50:23.162624+00:00"},{"alias_kind":"pith_short_16","alias_value":"5QYSIGSCYI7RAB46","created_at":"2026-07-05T00:50:23.162624+00:00"},{"alias_kind":"pith_short_8","alias_value":"5QYSIGSC","created_at":"2026-07-05T00:50:23.162624+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2106.09636","citing_title":"Multi-Stage Prototype Learning for Interpretable Time Series Classification","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5QYSIGSCYI7RAB46PKENETTT65","json":"https://pith.science/pith/5QYSIGSCYI7RAB46PKENETTT65.json","graph_json":"https://pith.science/api/pith-number/5QYSIGSCYI7RAB46PKENETTT65/graph.json","events_json":"https://pith.science/api/pith-number/5QYSIGSCYI7RAB46PKENETTT65/events.json","paper":"https://pith.science/paper/5QYSIGSC"},"agent_actions":{"view_html":"https://pith.science/pith/5QYSIGSCYI7RAB46PKENETTT65","download_json":"https://pith.science/pith/5QYSIGSCYI7RAB46PKENETTT65.json","view_paper":"https://pith.science/paper/5QYSIGSC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1705.07356&json=true","fetch_graph":"https://pith.science/api/pith-number/5QYSIGSCYI7RAB46PKENETTT65/graph.json","fetch_events":"https://pith.science/api/pith-number/5QYSIGSCYI7RAB46PKENETTT65/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5QYSIGSCYI7RAB46PKENETTT65/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5QYSIGSCYI7RAB46PKENETTT65/action/storage_attestation","attest_author":"https://pith.science/pith/5QYSIGSCYI7RAB46PKENETTT65/action/author_attestation","sign_citation":"https://pith.science/pith/5QYSIGSCYI7RAB46PKENETTT65/action/citation_signature","submit_replication":"https://pith.science/pith/5QYSIGSCYI7RAB46PKENETTT65/action/replication_record"}},"created_at":"2026-07-05T00:50:23.162624+00:00","updated_at":"2026-07-05T00:50:23.162624+00:00"}