{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:YICQWWKHHLJ7EFH653X3NXC273","short_pith_number":"pith:YICQWWKH","canonical_record":{"source":{"id":"1709.06247","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2017-09-19T04:27:56Z","cross_cats_sorted":[],"title_canon_sha256":"17e517543c37452f65184fbc2685897954a665852ab4b99b503d50839bbb653e","abstract_canon_sha256":"47891af4f96624a83c41793ac70a29c77fe072d28d0e8b2f82e42d41df5b5d6a"},"schema_version":"1.0"},"canonical_sha256":"c2050b59473ad3f214feeeefb6dc5afef62287aec0cb280f3ea085aa17e310ee","source":{"kind":"arxiv","id":"1709.06247","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1709.06247","created_at":"2026-05-18T00:06:48Z"},{"alias_kind":"arxiv_version","alias_value":"1709.06247v2","created_at":"2026-05-18T00:06:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1709.06247","created_at":"2026-05-18T00:06:48Z"},{"alias_kind":"pith_short_12","alias_value":"YICQWWKHHLJ7","created_at":"2026-05-18T12:31:56Z"},{"alias_kind":"pith_short_16","alias_value":"YICQWWKHHLJ7EFH6","created_at":"2026-05-18T12:31:56Z"},{"alias_kind":"pith_short_8","alias_value":"YICQWWKH","created_at":"2026-05-18T12:31:56Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:YICQWWKHHLJ7EFH653X3NXC273","target":"record","payload":{"canonical_record":{"source":{"id":"1709.06247","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2017-09-19T04:27:56Z","cross_cats_sorted":[],"title_canon_sha256":"17e517543c37452f65184fbc2685897954a665852ab4b99b503d50839bbb653e","abstract_canon_sha256":"47891af4f96624a83c41793ac70a29c77fe072d28d0e8b2f82e42d41df5b5d6a"},"schema_version":"1.0"},"canonical_sha256":"c2050b59473ad3f214feeeefb6dc5afef62287aec0cb280f3ea085aa17e310ee","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:06:48.392686Z","signature_b64":"T9xmeAJGEeuv4FYtrmyriVddspYQzQESDNWQHxlt5DZkRCLSu5N+7ejd2aNVrS9RRtuixHdfA6Rg4JzBTPuEDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c2050b59473ad3f214feeeefb6dc5afef62287aec0cb280f3ea085aa17e310ee","last_reissued_at":"2026-05-18T00:06:48.392207Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:06:48.392207Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1709.06247","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":"2026-05-18T00:06:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fopxfwKlQ2PYkh1FI07Wb2bS0tOwnDdI5h6ABj8XjmGoJWsdlayucpZ0Ccm9VN27AaeKGcRnTtvK1tzMhImTDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T05:49:35.310760Z"},"content_sha256":"d40e1510b9e286f95faf1db1c903817eb17a126edf16650beab695a92f2905b3","schema_version":"1.0","event_id":"sha256:d40e1510b9e286f95faf1db1c903817eb17a126edf16650beab695a92f2905b3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:YICQWWKHHLJ7EFH653X3NXC273","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Rethink ReLU to Training Better CNNs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Gangming Zhao, He Guan, Jingdong Wang, Peng Tang, Zhaoxiang Zhang","submitted_at":"2017-09-19T04:27:56Z","abstract_excerpt":"Most of convolutional neural networks share the same characteristic: each convolutional layer is followed by a nonlinear activation layer where Rectified Linear Unit (ReLU) is the most widely used. In this paper, we argue that the designed structure with the equal ratio between these two layers may not be the best choice since it could result in the poor generalization ability. Thus, we try to investigate a more suitable method on using ReLU to explore the better network architectures. Specifically, we propose a proportional module to keep the ratio between convolution and ReLU amount to be N:"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1709.06247","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":""},"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":"2026-05-18T00:06:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ErHft1aVYyp2L7ihRCTwR+r0prS2zynl0/TZyb6EjEOjtJY2G/h4wBf9wiE6mcgmxFz3sTWQeiZDHNObHEG4Dw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T05:49:35.311418Z"},"content_sha256":"687881e1d843d5837991c49b84f0c2ad89f9c709e159d601f94d980660dcc09c","schema_version":"1.0","event_id":"sha256:687881e1d843d5837991c49b84f0c2ad89f9c709e159d601f94d980660dcc09c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YICQWWKHHLJ7EFH653X3NXC273/bundle.json","state_url":"https://pith.science/pith/YICQWWKHHLJ7EFH653X3NXC273/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YICQWWKHHLJ7EFH653X3NXC273/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":"2026-05-31T05:49:35Z","links":{"resolver":"https://pith.science/pith/YICQWWKHHLJ7EFH653X3NXC273","bundle":"https://pith.science/pith/YICQWWKHHLJ7EFH653X3NXC273/bundle.json","state":"https://pith.science/pith/YICQWWKHHLJ7EFH653X3NXC273/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YICQWWKHHLJ7EFH653X3NXC273/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:YICQWWKHHLJ7EFH653X3NXC273","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":"47891af4f96624a83c41793ac70a29c77fe072d28d0e8b2f82e42d41df5b5d6a","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2017-09-19T04:27:56Z","title_canon_sha256":"17e517543c37452f65184fbc2685897954a665852ab4b99b503d50839bbb653e"},"schema_version":"1.0","source":{"id":"1709.06247","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1709.06247","created_at":"2026-05-18T00:06:48Z"},{"alias_kind":"arxiv_version","alias_value":"1709.06247v2","created_at":"2026-05-18T00:06:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1709.06247","created_at":"2026-05-18T00:06:48Z"},{"alias_kind":"pith_short_12","alias_value":"YICQWWKHHLJ7","created_at":"2026-05-18T12:31:56Z"},{"alias_kind":"pith_short_16","alias_value":"YICQWWKHHLJ7EFH6","created_at":"2026-05-18T12:31:56Z"},{"alias_kind":"pith_short_8","alias_value":"YICQWWKH","created_at":"2026-05-18T12:31:56Z"}],"graph_snapshots":[{"event_id":"sha256:687881e1d843d5837991c49b84f0c2ad89f9c709e159d601f94d980660dcc09c","target":"graph","created_at":"2026-05-18T00:06:48Z","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"},"paper":{"abstract_excerpt":"Most of convolutional neural networks share the same characteristic: each convolutional layer is followed by a nonlinear activation layer where Rectified Linear Unit (ReLU) is the most widely used. In this paper, we argue that the designed structure with the equal ratio between these two layers may not be the best choice since it could result in the poor generalization ability. Thus, we try to investigate a more suitable method on using ReLU to explore the better network architectures. Specifically, we propose a proportional module to keep the ratio between convolution and ReLU amount to be N:","authors_text":"Gangming Zhao, He Guan, Jingdong Wang, Peng Tang, Zhaoxiang Zhang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2017-09-19T04:27:56Z","title":"Rethink ReLU to Training Better CNNs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1709.06247","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:d40e1510b9e286f95faf1db1c903817eb17a126edf16650beab695a92f2905b3","target":"record","created_at":"2026-05-18T00:06:48Z","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":"47891af4f96624a83c41793ac70a29c77fe072d28d0e8b2f82e42d41df5b5d6a","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2017-09-19T04:27:56Z","title_canon_sha256":"17e517543c37452f65184fbc2685897954a665852ab4b99b503d50839bbb653e"},"schema_version":"1.0","source":{"id":"1709.06247","kind":"arxiv","version":2}},"canonical_sha256":"c2050b59473ad3f214feeeefb6dc5afef62287aec0cb280f3ea085aa17e310ee","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c2050b59473ad3f214feeeefb6dc5afef62287aec0cb280f3ea085aa17e310ee","first_computed_at":"2026-05-18T00:06:48.392207Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:06:48.392207Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"T9xmeAJGEeuv4FYtrmyriVddspYQzQESDNWQHxlt5DZkRCLSu5N+7ejd2aNVrS9RRtuixHdfA6Rg4JzBTPuEDg==","signature_status":"signed_v1","signed_at":"2026-05-18T00:06:48.392686Z","signed_message":"canonical_sha256_bytes"},"source_id":"1709.06247","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d40e1510b9e286f95faf1db1c903817eb17a126edf16650beab695a92f2905b3","sha256:687881e1d843d5837991c49b84f0c2ad89f9c709e159d601f94d980660dcc09c"],"state_sha256":"7cf169cf81d9f5ec67254553edc44a9c255cab3949982238618f3806c59238e6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"17oUPq2wPt/vPYH2Hx2XD5Ix56Il1j5182w8yX6cPs6DTatu01+8SBslfZGC6TC/0jOATC1S7Jo7tLakHpdQCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-31T05:49:35.315298Z","bundle_sha256":"db8b6c03cd33cc7f8687cd2cc02270332eea0e457ab9b782da7b0266d3fdb548"}}