{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:7S52UX24ZZVHJKCFQX5OQN3N2Z","short_pith_number":"pith:7S52UX24","canonical_record":{"source":{"id":"2105.02498","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-05-06T08:03:45Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"d384fdadb7514c7c504da1ddc3f120fb44ba5a7378db85d3453e30cccb626eb0","abstract_canon_sha256":"cc208de1e6fb2070b6073f95eaf507bcaa9f52fb87c31630f628d2812d19a6ae"},"schema_version":"1.0"},"canonical_sha256":"fcbbaa5f5cce6a74a84585fae8376dd66320f19000ded4f37b4b2028d0d633dc","source":{"kind":"arxiv","id":"2105.02498","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2105.02498","created_at":"2026-07-05T03:00:09Z"},{"alias_kind":"arxiv_version","alias_value":"2105.02498v2","created_at":"2026-07-05T03:00:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.02498","created_at":"2026-07-05T03:00:09Z"},{"alias_kind":"pith_short_12","alias_value":"7S52UX24ZZVH","created_at":"2026-07-05T03:00:09Z"},{"alias_kind":"pith_short_16","alias_value":"7S52UX24ZZVHJKCF","created_at":"2026-07-05T03:00:09Z"},{"alias_kind":"pith_short_8","alias_value":"7S52UX24","created_at":"2026-07-05T03:00:09Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:7S52UX24ZZVHJKCFQX5OQN3N2Z","target":"record","payload":{"canonical_record":{"source":{"id":"2105.02498","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-05-06T08:03:45Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"d384fdadb7514c7c504da1ddc3f120fb44ba5a7378db85d3453e30cccb626eb0","abstract_canon_sha256":"cc208de1e6fb2070b6073f95eaf507bcaa9f52fb87c31630f628d2812d19a6ae"},"schema_version":"1.0"},"canonical_sha256":"fcbbaa5f5cce6a74a84585fae8376dd66320f19000ded4f37b4b2028d0d633dc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:00:09.310229Z","signature_b64":"497UHp1O9CVmUoZGT+Fa3ZAJJCZr4yzsJO1iRDu4raGzg86PdJNFxxvbMb7HoMEx9xPgcPS7BKRU+GQXsypxBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fcbbaa5f5cce6a74a84585fae8376dd66320f19000ded4f37b4b2028d0d633dc","last_reissued_at":"2026-07-05T03:00:09.309851Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:00:09.309851Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2105.02498","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T03:00:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gxZR44NYaMo3DoplXg6at5CyeQBjUKIq2r28jabcNZTwha9FJ3AmNwlDB13/YC9NLwCMHjI1Hh8wqGGW4OndCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T04:52:59.193777Z"},"content_sha256":"70eeae9393367b0b514ff426f8bcac34d47662a59e629291308bd479522606d1","schema_version":"1.0","event_id":"sha256:70eeae9393367b0b514ff426f8bcac34d47662a59e629291308bd479522606d1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:7S52UX24ZZVHJKCFQX5OQN3N2Z","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Why Approximate Matrix Square Root Outperforms Accurate SVD in Global Covariance Pooling?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Nicu Sebe, Wei Wang, Yue Song","submitted_at":"2021-05-06T08:03:45Z","abstract_excerpt":"Global covariance pooling (GCP) aims at exploiting the second-order statistics of the convolutional feature. Its effectiveness has been demonstrated in boosting the classification performance of Convolutional Neural Networks (CNNs). Singular Value Decomposition (SVD) is used in GCP to compute the matrix square root. However, the approximate matrix square root calculated using Newton-Schulz iteration \\cite{li2018towards} outperforms the accurate one computed via SVD \\cite{li2017second}. We empirically analyze the reason behind the performance gap from the perspectives of data precision and grad"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.02498","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/2105.02498/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T03:00:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Tp3bMrbIwXt8B6GJG67+AVZVISL2fZ1QkTa/UHZakgmx6AT9dOIzT9d4r/BHhZC48lnH0xvdcNq3W/9FReoiBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T04:52:59.194300Z"},"content_sha256":"4f44a828d4c6fa14b8168706914a3bfebbda5df4cace1c7bde082407a29b0530","schema_version":"1.0","event_id":"sha256:4f44a828d4c6fa14b8168706914a3bfebbda5df4cace1c7bde082407a29b0530"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7S52UX24ZZVHJKCFQX5OQN3N2Z/bundle.json","state_url":"https://pith.science/pith/7S52UX24ZZVHJKCFQX5OQN3N2Z/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7S52UX24ZZVHJKCFQX5OQN3N2Z/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-09T04:52:59Z","links":{"resolver":"https://pith.science/pith/7S52UX24ZZVHJKCFQX5OQN3N2Z","bundle":"https://pith.science/pith/7S52UX24ZZVHJKCFQX5OQN3N2Z/bundle.json","state":"https://pith.science/pith/7S52UX24ZZVHJKCFQX5OQN3N2Z/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7S52UX24ZZVHJKCFQX5OQN3N2Z/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:7S52UX24ZZVHJKCFQX5OQN3N2Z","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"cc208de1e6fb2070b6073f95eaf507bcaa9f52fb87c31630f628d2812d19a6ae","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-05-06T08:03:45Z","title_canon_sha256":"d384fdadb7514c7c504da1ddc3f120fb44ba5a7378db85d3453e30cccb626eb0"},"schema_version":"1.0","source":{"id":"2105.02498","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2105.02498","created_at":"2026-07-05T03:00:09Z"},{"alias_kind":"arxiv_version","alias_value":"2105.02498v2","created_at":"2026-07-05T03:00:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.02498","created_at":"2026-07-05T03:00:09Z"},{"alias_kind":"pith_short_12","alias_value":"7S52UX24ZZVH","created_at":"2026-07-05T03:00:09Z"},{"alias_kind":"pith_short_16","alias_value":"7S52UX24ZZVHJKCF","created_at":"2026-07-05T03:00:09Z"},{"alias_kind":"pith_short_8","alias_value":"7S52UX24","created_at":"2026-07-05T03:00:09Z"}],"graph_snapshots":[{"event_id":"sha256:4f44a828d4c6fa14b8168706914a3bfebbda5df4cace1c7bde082407a29b0530","target":"graph","created_at":"2026-07-05T03:00:09Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2105.02498/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Global covariance pooling (GCP) aims at exploiting the second-order statistics of the convolutional feature. Its effectiveness has been demonstrated in boosting the classification performance of Convolutional Neural Networks (CNNs). Singular Value Decomposition (SVD) is used in GCP to compute the matrix square root. However, the approximate matrix square root calculated using Newton-Schulz iteration \\cite{li2018towards} outperforms the accurate one computed via SVD \\cite{li2017second}. We empirically analyze the reason behind the performance gap from the perspectives of data precision and grad","authors_text":"Nicu Sebe, Wei Wang, Yue Song","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-05-06T08:03:45Z","title":"Why Approximate Matrix Square Root Outperforms Accurate SVD in Global Covariance Pooling?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.02498","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:70eeae9393367b0b514ff426f8bcac34d47662a59e629291308bd479522606d1","target":"record","created_at":"2026-07-05T03:00:09Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"cc208de1e6fb2070b6073f95eaf507bcaa9f52fb87c31630f628d2812d19a6ae","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-05-06T08:03:45Z","title_canon_sha256":"d384fdadb7514c7c504da1ddc3f120fb44ba5a7378db85d3453e30cccb626eb0"},"schema_version":"1.0","source":{"id":"2105.02498","kind":"arxiv","version":2}},"canonical_sha256":"fcbbaa5f5cce6a74a84585fae8376dd66320f19000ded4f37b4b2028d0d633dc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fcbbaa5f5cce6a74a84585fae8376dd66320f19000ded4f37b4b2028d0d633dc","first_computed_at":"2026-07-05T03:00:09.309851Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:00:09.309851Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"497UHp1O9CVmUoZGT+Fa3ZAJJCZr4yzsJO1iRDu4raGzg86PdJNFxxvbMb7HoMEx9xPgcPS7BKRU+GQXsypxBw==","signature_status":"signed_v1","signed_at":"2026-07-05T03:00:09.310229Z","signed_message":"canonical_sha256_bytes"},"source_id":"2105.02498","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:70eeae9393367b0b514ff426f8bcac34d47662a59e629291308bd479522606d1","sha256:4f44a828d4c6fa14b8168706914a3bfebbda5df4cace1c7bde082407a29b0530"],"state_sha256":"c195c7f4245512d2531ce3982699e62130ecc77a276a5466a5140058daa11b69"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gr+YPE5u4kGDB55JyH1/EzbeC6qIlUxMUdPQgBwlwhICG0w4h2bM3xr+SVNZ5ZBEBZLUKpZJ7gkgdfHiiC60Dw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T04:52:59.198444Z","bundle_sha256":"f0fd59066d9fcdee0fac52e77c95004aeca0dc0e461e688e7a46c9fda9930e2e"}}