{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:ACCHRQ2HUUOWMGHAWJN55GRSJG","short_pith_number":"pith:ACCHRQ2H","schema_version":"1.0","canonical_sha256":"008478c347a51d6618e0b25bde9a32498a647d874e10ae55daef0d800fa43e42","source":{"kind":"arxiv","id":"2301.12033","version":1},"attestation_state":"computed","paper":{"title":"Norm-based Generalization Bounds for Compositionally Sparse Neural Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Liane Galanti, Mengjia Xu, Tomaso Poggio, Tomer Galanti","submitted_at":"2023-01-28T00:06:22Z","abstract_excerpt":"In this paper, we investigate the Rademacher complexity of deep sparse neural networks, where each neuron receives a small number of inputs. We prove generalization bounds for multilayered sparse ReLU neural networks, including convolutional neural networks. These bounds differ from previous ones, as they consider the norms of the convolutional filters instead of the norms of the associated Toeplitz matrices, independently of weight sharing between neurons.\n  As we show theoretically, these bounds may be orders of magnitude better than standard norm-based generalization bounds and empirically,"},"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":"2301.12033","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-01-28T00:06:22Z","cross_cats_sorted":[],"title_canon_sha256":"18e4587f5ae2528378cc273822babb0b5451bc4cc623d7b5a980f4fc6a57fb62","abstract_canon_sha256":"7e690111b9eef684bcd70346ba73251312bdcaa66994ae8ec0484a4f97541490"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:36:34.486575Z","signature_b64":"1YeGHsJYBuCiNW+3t331tX8+43IZLey8kvl6chyyyTtglrZgs2/+btb/I/ivnmpPxdS1LgoNa0OKstGBclqfDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"008478c347a51d6618e0b25bde9a32498a647d874e10ae55daef0d800fa43e42","last_reissued_at":"2026-07-05T05:36:34.486113Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:36:34.486113Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Norm-based Generalization Bounds for Compositionally Sparse Neural Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Liane Galanti, Mengjia Xu, Tomaso Poggio, Tomer Galanti","submitted_at":"2023-01-28T00:06:22Z","abstract_excerpt":"In this paper, we investigate the Rademacher complexity of deep sparse neural networks, where each neuron receives a small number of inputs. We prove generalization bounds for multilayered sparse ReLU neural networks, including convolutional neural networks. These bounds differ from previous ones, as they consider the norms of the convolutional filters instead of the norms of the associated Toeplitz matrices, independently of weight sharing between neurons.\n  As we show theoretically, these bounds may be orders of magnitude better than standard norm-based generalization bounds and empirically,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.12033","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/2301.12033/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":"2301.12033","created_at":"2026-07-05T05:36:34.486169+00:00"},{"alias_kind":"arxiv_version","alias_value":"2301.12033v1","created_at":"2026-07-05T05:36:34.486169+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.12033","created_at":"2026-07-05T05:36:34.486169+00:00"},{"alias_kind":"pith_short_12","alias_value":"ACCHRQ2HUUOW","created_at":"2026-07-05T05:36:34.486169+00:00"},{"alias_kind":"pith_short_16","alias_value":"ACCHRQ2HUUOWMGHA","created_at":"2026-07-05T05:36:34.486169+00:00"},{"alias_kind":"pith_short_8","alias_value":"ACCHRQ2H","created_at":"2026-07-05T05:36:34.486169+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.25608","citing_title":"Learning Sparse Compositional Functions with Norm-Constrained Neural Networks","ref_index":32,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ACCHRQ2HUUOWMGHAWJN55GRSJG","json":"https://pith.science/pith/ACCHRQ2HUUOWMGHAWJN55GRSJG.json","graph_json":"https://pith.science/api/pith-number/ACCHRQ2HUUOWMGHAWJN55GRSJG/graph.json","events_json":"https://pith.science/api/pith-number/ACCHRQ2HUUOWMGHAWJN55GRSJG/events.json","paper":"https://pith.science/paper/ACCHRQ2H"},"agent_actions":{"view_html":"https://pith.science/pith/ACCHRQ2HUUOWMGHAWJN55GRSJG","download_json":"https://pith.science/pith/ACCHRQ2HUUOWMGHAWJN55GRSJG.json","view_paper":"https://pith.science/paper/ACCHRQ2H","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2301.12033&json=true","fetch_graph":"https://pith.science/api/pith-number/ACCHRQ2HUUOWMGHAWJN55GRSJG/graph.json","fetch_events":"https://pith.science/api/pith-number/ACCHRQ2HUUOWMGHAWJN55GRSJG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ACCHRQ2HUUOWMGHAWJN55GRSJG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ACCHRQ2HUUOWMGHAWJN55GRSJG/action/storage_attestation","attest_author":"https://pith.science/pith/ACCHRQ2HUUOWMGHAWJN55GRSJG/action/author_attestation","sign_citation":"https://pith.science/pith/ACCHRQ2HUUOWMGHAWJN55GRSJG/action/citation_signature","submit_replication":"https://pith.science/pith/ACCHRQ2HUUOWMGHAWJN55GRSJG/action/replication_record"}},"created_at":"2026-07-05T05:36:34.486169+00:00","updated_at":"2026-07-05T05:36:34.486169+00:00"}