{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:DO2R5MTFRULOTOYDADZH2I4L4G","short_pith_number":"pith:DO2R5MTF","canonical_record":{"source":{"id":"1901.06773","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-21T02:52:09Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"e6dd7a575ccb90ce866a44f875d0ea667e3ed16cf92f0b2c1c36a64d7988cc71","abstract_canon_sha256":"30668772f64f240f598cd932c6a002f69d4b3670ca7f25545bde897884f492e9"},"schema_version":"1.0"},"canonical_sha256":"1bb51eb2658d16e9bb0300f27d238be1814312e42577d00b96817ccef64952f0","source":{"kind":"arxiv","id":"1901.06773","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1901.06773","created_at":"2026-05-17T23:42:53Z"},{"alias_kind":"arxiv_version","alias_value":"1901.06773v2","created_at":"2026-05-17T23:42:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1901.06773","created_at":"2026-05-17T23:42:53Z"},{"alias_kind":"pith_short_12","alias_value":"DO2R5MTFRULO","created_at":"2026-05-18T12:33:15Z"},{"alias_kind":"pith_short_16","alias_value":"DO2R5MTFRULOTOYD","created_at":"2026-05-18T12:33:15Z"},{"alias_kind":"pith_short_8","alias_value":"DO2R5MTF","created_at":"2026-05-18T12:33:15Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:DO2R5MTFRULOTOYDADZH2I4L4G","target":"record","payload":{"canonical_record":{"source":{"id":"1901.06773","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-21T02:52:09Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"e6dd7a575ccb90ce866a44f875d0ea667e3ed16cf92f0b2c1c36a64d7988cc71","abstract_canon_sha256":"30668772f64f240f598cd932c6a002f69d4b3670ca7f25545bde897884f492e9"},"schema_version":"1.0"},"canonical_sha256":"1bb51eb2658d16e9bb0300f27d238be1814312e42577d00b96817ccef64952f0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:42:53.453039Z","signature_b64":"s1T/qiDSN7/cKFbcc1ByALLD1K+zH2prGHWwrz2AguvHbN+Sn9NoU3riVrTh1LGz8mWKwY2YFOF/YC0Enc34AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1bb51eb2658d16e9bb0300f27d238be1814312e42577d00b96817ccef64952f0","last_reissued_at":"2026-05-17T23:42:53.452371Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:42:53.452371Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1901.06773","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":"2026-05-17T23:42:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lpV7oAGv3uIvPCKw500cYn2EbpPD5vQaNITkRZqm+R1Nl7eGEkMg/66a/BAtnRl2wthR4BayMBBA2sSxqxOSDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T23:53:46.147993Z"},"content_sha256":"b6983f460040ca2fe16776ee43171fb3124cecae7ead4ec7216c917d636f273a","schema_version":"1.0","event_id":"sha256:b6983f460040ca2fe16776ee43171fb3124cecae7ead4ec7216c917d636f273a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:DO2R5MTFRULOTOYDADZH2I4L4G","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"AccUDNN: A GPU Memory Efficient Accelerator for Training Ultra-deep Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"jinrong Guo, Jizhong Han, Qu Lu, Ruixuan Li, Songlin Hu, Wang Wang, Wantao Liu","submitted_at":"2019-01-21T02:52:09Z","abstract_excerpt":"Typically, Ultra-deep neural network(UDNN) tends to yield high-quality model, but its training process is usually resource intensive and time-consuming. Modern GPU's scarce DRAM capacity is the primary bottleneck that hinders the trainability and the training efficiency of UDNN. In this paper, we present \"AccUDNN\", an accelerator that aims to make the utmost use of finite GPU memory resources to speed up the training process of UDNN. AccUDNN mainly includes two modules: memory optimizer and hyperparameter tuner. Memory optimizer develops a performance-model guided dynamic swap out/in strategy,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1901.06773","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":"2026-05-17T23:42:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FTyynErDjaISw/LikL2IndO5LIqdfaoRdcbW2mKkJBKiJGm4kTEA0vrcg16DNzBv/PCuoUMBsEMb1T9Cqx3/AA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T23:53:46.148805Z"},"content_sha256":"81fc2dccab50b71b4c7fb06a6a03bf0c33f7fbd887a1791b98baed9717b7be42","schema_version":"1.0","event_id":"sha256:81fc2dccab50b71b4c7fb06a6a03bf0c33f7fbd887a1791b98baed9717b7be42"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DO2R5MTFRULOTOYDADZH2I4L4G/bundle.json","state_url":"https://pith.science/pith/DO2R5MTFRULOTOYDADZH2I4L4G/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DO2R5MTFRULOTOYDADZH2I4L4G/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":"2026-05-31T23:53:46Z","links":{"resolver":"https://pith.science/pith/DO2R5MTFRULOTOYDADZH2I4L4G","bundle":"https://pith.science/pith/DO2R5MTFRULOTOYDADZH2I4L4G/bundle.json","state":"https://pith.science/pith/DO2R5MTFRULOTOYDADZH2I4L4G/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DO2R5MTFRULOTOYDADZH2I4L4G/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:DO2R5MTFRULOTOYDADZH2I4L4G","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":"30668772f64f240f598cd932c6a002f69d4b3670ca7f25545bde897884f492e9","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-21T02:52:09Z","title_canon_sha256":"e6dd7a575ccb90ce866a44f875d0ea667e3ed16cf92f0b2c1c36a64d7988cc71"},"schema_version":"1.0","source":{"id":"1901.06773","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1901.06773","created_at":"2026-05-17T23:42:53Z"},{"alias_kind":"arxiv_version","alias_value":"1901.06773v2","created_at":"2026-05-17T23:42:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1901.06773","created_at":"2026-05-17T23:42:53Z"},{"alias_kind":"pith_short_12","alias_value":"DO2R5MTFRULO","created_at":"2026-05-18T12:33:15Z"},{"alias_kind":"pith_short_16","alias_value":"DO2R5MTFRULOTOYD","created_at":"2026-05-18T12:33:15Z"},{"alias_kind":"pith_short_8","alias_value":"DO2R5MTF","created_at":"2026-05-18T12:33:15Z"}],"graph_snapshots":[{"event_id":"sha256:81fc2dccab50b71b4c7fb06a6a03bf0c33f7fbd887a1791b98baed9717b7be42","target":"graph","created_at":"2026-05-17T23:42:53Z","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":"Typically, Ultra-deep neural network(UDNN) tends to yield high-quality model, but its training process is usually resource intensive and time-consuming. Modern GPU's scarce DRAM capacity is the primary bottleneck that hinders the trainability and the training efficiency of UDNN. In this paper, we present \"AccUDNN\", an accelerator that aims to make the utmost use of finite GPU memory resources to speed up the training process of UDNN. AccUDNN mainly includes two modules: memory optimizer and hyperparameter tuner. Memory optimizer develops a performance-model guided dynamic swap out/in strategy,","authors_text":"jinrong Guo, Jizhong Han, Qu Lu, Ruixuan Li, Songlin Hu, Wang Wang, Wantao Liu","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-21T02:52:09Z","title":"AccUDNN: A GPU Memory Efficient Accelerator for Training Ultra-deep Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1901.06773","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:b6983f460040ca2fe16776ee43171fb3124cecae7ead4ec7216c917d636f273a","target":"record","created_at":"2026-05-17T23:42:53Z","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":"30668772f64f240f598cd932c6a002f69d4b3670ca7f25545bde897884f492e9","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-21T02:52:09Z","title_canon_sha256":"e6dd7a575ccb90ce866a44f875d0ea667e3ed16cf92f0b2c1c36a64d7988cc71"},"schema_version":"1.0","source":{"id":"1901.06773","kind":"arxiv","version":2}},"canonical_sha256":"1bb51eb2658d16e9bb0300f27d238be1814312e42577d00b96817ccef64952f0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1bb51eb2658d16e9bb0300f27d238be1814312e42577d00b96817ccef64952f0","first_computed_at":"2026-05-17T23:42:53.452371Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:42:53.452371Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"s1T/qiDSN7/cKFbcc1ByALLD1K+zH2prGHWwrz2AguvHbN+Sn9NoU3riVrTh1LGz8mWKwY2YFOF/YC0Enc34AQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:42:53.453039Z","signed_message":"canonical_sha256_bytes"},"source_id":"1901.06773","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b6983f460040ca2fe16776ee43171fb3124cecae7ead4ec7216c917d636f273a","sha256:81fc2dccab50b71b4c7fb06a6a03bf0c33f7fbd887a1791b98baed9717b7be42"],"state_sha256":"e15b07481868597d04dfb814d714f489f9b8670a5609b6075e73ac7960ce03b8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lXqSLW1xknrGwEfw5jh9ybLNX1vSBe2CLiTgAbw9pW5Cr7/4nNzksFGRYQBwsGP7Sl867ImXP/Iuj3GvN+OQCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-31T23:53:46.153291Z","bundle_sha256":"289baf8ffcf2613dbaef7b0d2ad0bbeeb2854aa41b3411697ab789269be5973c"}}