{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:NSALULH6272NFM7WTV4SWQFGC6","short_pith_number":"pith:NSALULH6","schema_version":"1.0","canonical_sha256":"6c80ba2cfed7f4d2b3f69d792b40a617b90531af091e8259549ae03f2f473a85","source":{"kind":"arxiv","id":"2202.01653","version":1},"attestation_state":"computed","paper":{"title":"Learning strides in convolutional neural networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"David Grangier, Neil Zeghidour, Olivier Teboul, Rachid Riad","submitted_at":"2022-02-03T16:03:36Z","abstract_excerpt":"Convolutional neural networks typically contain several downsampling operators, such as strided convolutions or pooling layers, that progressively reduce the resolution of intermediate representations. This provides some shift-invariance while reducing the computational complexity of the whole architecture. A critical hyperparameter of such layers is their stride: the integer factor of downsampling. As strides are not differentiable, finding the best configuration either requires cross-validation or discrete optimization (e.g. architecture search), which rapidly become prohibitive as the searc"},"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":"2202.01653","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-02-03T16:03:36Z","cross_cats_sorted":[],"title_canon_sha256":"86a5abccac4e865916eff9034dd4545cec9995e28b0c56e88462884df2426fd3","abstract_canon_sha256":"4d5d37f439ffaf439a9f3cc06b67d8d1e3a2c3a7e97970115f55bdf3238fa8d1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:53:56.576552Z","signature_b64":"zff+EOv8Hfto6stDOaQ+JUtGNnvno+uovhxl+81iSpy2Cm7j74sSJacEos8HPcYK9w8sLp1McWqMTx9Zd1CcCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6c80ba2cfed7f4d2b3f69d792b40a617b90531af091e8259549ae03f2f473a85","last_reissued_at":"2026-07-05T03:53:56.576028Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:53:56.576028Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning strides in convolutional neural networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"David Grangier, Neil Zeghidour, Olivier Teboul, Rachid Riad","submitted_at":"2022-02-03T16:03:36Z","abstract_excerpt":"Convolutional neural networks typically contain several downsampling operators, such as strided convolutions or pooling layers, that progressively reduce the resolution of intermediate representations. This provides some shift-invariance while reducing the computational complexity of the whole architecture. A critical hyperparameter of such layers is their stride: the integer factor of downsampling. As strides are not differentiable, finding the best configuration either requires cross-validation or discrete optimization (e.g. architecture search), which rapidly become prohibitive as the searc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2202.01653","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/2202.01653/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":"2202.01653","created_at":"2026-07-05T03:53:56.576089+00:00"},{"alias_kind":"arxiv_version","alias_value":"2202.01653v1","created_at":"2026-07-05T03:53:56.576089+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2202.01653","created_at":"2026-07-05T03:53:56.576089+00:00"},{"alias_kind":"pith_short_12","alias_value":"NSALULH6272N","created_at":"2026-07-05T03:53:56.576089+00:00"},{"alias_kind":"pith_short_16","alias_value":"NSALULH6272NFM7W","created_at":"2026-07-05T03:53:56.576089+00:00"},{"alias_kind":"pith_short_8","alias_value":"NSALULH6","created_at":"2026-07-05T03:53:56.576089+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.18113","citing_title":"Beyond Discreteness: Sample Complexity Analysis of Straight-Through Estimator for 1-bit Quantization","ref_index":41,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NSALULH6272NFM7WTV4SWQFGC6","json":"https://pith.science/pith/NSALULH6272NFM7WTV4SWQFGC6.json","graph_json":"https://pith.science/api/pith-number/NSALULH6272NFM7WTV4SWQFGC6/graph.json","events_json":"https://pith.science/api/pith-number/NSALULH6272NFM7WTV4SWQFGC6/events.json","paper":"https://pith.science/paper/NSALULH6"},"agent_actions":{"view_html":"https://pith.science/pith/NSALULH6272NFM7WTV4SWQFGC6","download_json":"https://pith.science/pith/NSALULH6272NFM7WTV4SWQFGC6.json","view_paper":"https://pith.science/paper/NSALULH6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2202.01653&json=true","fetch_graph":"https://pith.science/api/pith-number/NSALULH6272NFM7WTV4SWQFGC6/graph.json","fetch_events":"https://pith.science/api/pith-number/NSALULH6272NFM7WTV4SWQFGC6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NSALULH6272NFM7WTV4SWQFGC6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NSALULH6272NFM7WTV4SWQFGC6/action/storage_attestation","attest_author":"https://pith.science/pith/NSALULH6272NFM7WTV4SWQFGC6/action/author_attestation","sign_citation":"https://pith.science/pith/NSALULH6272NFM7WTV4SWQFGC6/action/citation_signature","submit_replication":"https://pith.science/pith/NSALULH6272NFM7WTV4SWQFGC6/action/replication_record"}},"created_at":"2026-07-05T03:53:56.576089+00:00","updated_at":"2026-07-05T03:53:56.576089+00:00"}