{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:E6H727AB2R6ALM4AIR6EO4CRD4","short_pith_number":"pith:E6H727AB","canonical_record":{"source":{"id":"1705.01626","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-05-03T21:07:47Z","cross_cats_sorted":["cs.AR"],"title_canon_sha256":"11a9fa296a886f17d9ace41ae9ea265f2415737ba18deed4be629036f2dc5d1d","abstract_canon_sha256":"fc25229923dc55429018f6048f3103655d26708b19bafd61ac800324acb08c52"},"schema_version":"1.0"},"canonical_sha256":"278ffd7c01d47c05b380447c4770511f045d54f36cc43f813333d1a9e12e002c","source":{"kind":"arxiv","id":"1705.01626","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1705.01626","created_at":"2026-05-18T00:45:04Z"},{"alias_kind":"arxiv_version","alias_value":"1705.01626v1","created_at":"2026-05-18T00:45:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1705.01626","created_at":"2026-05-18T00:45:04Z"},{"alias_kind":"pith_short_12","alias_value":"E6H727AB2R6A","created_at":"2026-05-18T12:31:12Z"},{"alias_kind":"pith_short_16","alias_value":"E6H727AB2R6ALM4A","created_at":"2026-05-18T12:31:12Z"},{"alias_kind":"pith_short_8","alias_value":"E6H727AB","created_at":"2026-05-18T12:31:12Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:E6H727AB2R6ALM4AIR6EO4CRD4","target":"record","payload":{"canonical_record":{"source":{"id":"1705.01626","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-05-03T21:07:47Z","cross_cats_sorted":["cs.AR"],"title_canon_sha256":"11a9fa296a886f17d9ace41ae9ea265f2415737ba18deed4be629036f2dc5d1d","abstract_canon_sha256":"fc25229923dc55429018f6048f3103655d26708b19bafd61ac800324acb08c52"},"schema_version":"1.0"},"canonical_sha256":"278ffd7c01d47c05b380447c4770511f045d54f36cc43f813333d1a9e12e002c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:45:04.534691Z","signature_b64":"pU/GA/PIx8gdbyOWTK5uyqwgyema2oYM837rt1jzaCfZvQ6At7en7qD3aPYAEptabbJcoYoWi7gY+QExIeAPDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"278ffd7c01d47c05b380447c4770511f045d54f36cc43f813333d1a9e12e002c","last_reissued_at":"2026-05-18T00:45:04.534308Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:45:04.534308Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1705.01626","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:45:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3rCsy5kxjlOayuy8PHG0o/NJEyYI0gANQY3Ih0munmYi+3dwGGIb13wrflnMoMWjB6sN7+CNnAFSinsJSTVbAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T05:28:47.248223Z"},"content_sha256":"2c319362eaac1dafb1d5a0804f04dd9ef01d7ed529f09a3030d57bc6922d5878","schema_version":"1.0","event_id":"sha256:2c319362eaac1dafb1d5a0804f04dd9ef01d7ed529f09a3030d57bc6922d5878"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:E6H727AB2R6ALM4AIR6EO4CRD4","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Compressing DMA Engine: Leveraging Activation Sparsity for Training Deep Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AR"],"primary_cat":"cs.LG","authors_text":"Jeff Pool, Mike O'Connor, Minsoo Rhu, Niladrish Chatterjee, Stephen W. Keckler","submitted_at":"2017-05-03T21:07:47Z","abstract_excerpt":"Popular deep learning frameworks require users to fine-tune their memory usage so that the training data of a deep neural network (DNN) fits within the GPU physical memory. Prior work tries to address this restriction by virtualizing the memory usage of DNNs, enabling both CPU and GPU memory to be utilized for memory allocations. Despite its merits, virtualizing memory can incur significant performance overheads when the time needed to copy data back and forth from CPU memory is higher than the latency to perform the computations required for DNN forward and backward propagation. We introduce "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1705.01626","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":""},"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":"2026-05-18T00:45:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Ya+DaLrvV3+X+DYFG55KzMQ1eru84LQ9YYAmgc46Q98yhIAVgFpF409Lq0A6qOnw3nxdjMk7sa9OvyHpu2uNCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T05:28:47.248593Z"},"content_sha256":"a257b5e8872c2219e27ef4ca828024722e1eef5a3a081fe20d624ff6d30431bf","schema_version":"1.0","event_id":"sha256:a257b5e8872c2219e27ef4ca828024722e1eef5a3a081fe20d624ff6d30431bf"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/E6H727AB2R6ALM4AIR6EO4CRD4/bundle.json","state_url":"https://pith.science/pith/E6H727AB2R6ALM4AIR6EO4CRD4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/E6H727AB2R6ALM4AIR6EO4CRD4/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":"2026-05-30T05:28:47Z","links":{"resolver":"https://pith.science/pith/E6H727AB2R6ALM4AIR6EO4CRD4","bundle":"https://pith.science/pith/E6H727AB2R6ALM4AIR6EO4CRD4/bundle.json","state":"https://pith.science/pith/E6H727AB2R6ALM4AIR6EO4CRD4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/E6H727AB2R6ALM4AIR6EO4CRD4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:E6H727AB2R6ALM4AIR6EO4CRD4","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":"fc25229923dc55429018f6048f3103655d26708b19bafd61ac800324acb08c52","cross_cats_sorted":["cs.AR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-05-03T21:07:47Z","title_canon_sha256":"11a9fa296a886f17d9ace41ae9ea265f2415737ba18deed4be629036f2dc5d1d"},"schema_version":"1.0","source":{"id":"1705.01626","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1705.01626","created_at":"2026-05-18T00:45:04Z"},{"alias_kind":"arxiv_version","alias_value":"1705.01626v1","created_at":"2026-05-18T00:45:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1705.01626","created_at":"2026-05-18T00:45:04Z"},{"alias_kind":"pith_short_12","alias_value":"E6H727AB2R6A","created_at":"2026-05-18T12:31:12Z"},{"alias_kind":"pith_short_16","alias_value":"E6H727AB2R6ALM4A","created_at":"2026-05-18T12:31:12Z"},{"alias_kind":"pith_short_8","alias_value":"E6H727AB","created_at":"2026-05-18T12:31:12Z"}],"graph_snapshots":[{"event_id":"sha256:a257b5e8872c2219e27ef4ca828024722e1eef5a3a081fe20d624ff6d30431bf","target":"graph","created_at":"2026-05-18T00:45:04Z","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":"Popular deep learning frameworks require users to fine-tune their memory usage so that the training data of a deep neural network (DNN) fits within the GPU physical memory. Prior work tries to address this restriction by virtualizing the memory usage of DNNs, enabling both CPU and GPU memory to be utilized for memory allocations. Despite its merits, virtualizing memory can incur significant performance overheads when the time needed to copy data back and forth from CPU memory is higher than the latency to perform the computations required for DNN forward and backward propagation. We introduce ","authors_text":"Jeff Pool, Mike O'Connor, Minsoo Rhu, Niladrish Chatterjee, Stephen W. Keckler","cross_cats":["cs.AR"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-05-03T21:07:47Z","title":"Compressing DMA Engine: Leveraging Activation Sparsity for Training Deep Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1705.01626","kind":"arxiv","version":1},"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":"sha256:2c319362eaac1dafb1d5a0804f04dd9ef01d7ed529f09a3030d57bc6922d5878","target":"record","created_at":"2026-05-18T00:45:04Z","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":"fc25229923dc55429018f6048f3103655d26708b19bafd61ac800324acb08c52","cross_cats_sorted":["cs.AR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-05-03T21:07:47Z","title_canon_sha256":"11a9fa296a886f17d9ace41ae9ea265f2415737ba18deed4be629036f2dc5d1d"},"schema_version":"1.0","source":{"id":"1705.01626","kind":"arxiv","version":1}},"canonical_sha256":"278ffd7c01d47c05b380447c4770511f045d54f36cc43f813333d1a9e12e002c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"278ffd7c01d47c05b380447c4770511f045d54f36cc43f813333d1a9e12e002c","first_computed_at":"2026-05-18T00:45:04.534308Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:45:04.534308Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"pU/GA/PIx8gdbyOWTK5uyqwgyema2oYM837rt1jzaCfZvQ6At7en7qD3aPYAEptabbJcoYoWi7gY+QExIeAPDw==","signature_status":"signed_v1","signed_at":"2026-05-18T00:45:04.534691Z","signed_message":"canonical_sha256_bytes"},"source_id":"1705.01626","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2c319362eaac1dafb1d5a0804f04dd9ef01d7ed529f09a3030d57bc6922d5878","sha256:a257b5e8872c2219e27ef4ca828024722e1eef5a3a081fe20d624ff6d30431bf"],"state_sha256":"eaba04242800d8051f25c5a7d47240d86243bb77b8a877e72aa7afcad559222d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HTdcD1tL3nrXzbO7OHHPd7yk4di8FmkqtBd6qxtiQH9L8ewOkyDkMjotzwkDl3AIEjAKM9HAWsgdf5YGQQHMBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-30T05:28:47.250618Z","bundle_sha256":"e4690446fdd1fbd7b883e8dc51ddf41bccd0c405e126103fac7f671a6d382f93"}}