{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:DM6GBYSNJHDMWLODXZBVEPYJHJ","short_pith_number":"pith:DM6GBYSN","schema_version":"1.0","canonical_sha256":"1b3c60e24d49c6cb2dc3be43523f093a6ba54c54838dec733d4ba24e867e7264","source":{"kind":"arxiv","id":"1802.04924","version":2},"attestation_state":"computed","paper":{"title":"Exploring Hidden Dimensions in Parallelizing Convolutional Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC","cs.NE"],"primary_cat":"cs.LG","authors_text":"Alex Aiken, Charles R. Qi, Sina Lin, Zhihao Jia","submitted_at":"2018-02-14T02:00:40Z","abstract_excerpt":"The past few years have witnessed growth in the computational requirements for training deep convolutional neural networks. Current approaches parallelize training onto multiple devices by applying a single parallelization strategy (e.g., data or model parallelism) to all layers in a network. Although easy to reason about, these approaches result in suboptimal runtime performance in large-scale distributed training, since different layers in a network may prefer different parallelization strategies. In this paper, we propose layer-wise parallelism that allows each layer in a network to use an "},"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":"1802.04924","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-02-14T02:00:40Z","cross_cats_sorted":["cs.DC","cs.NE"],"title_canon_sha256":"1488af6627059c55eb10587c19d4c17566a1a40c8f972ff080453dbed4d7efe3","abstract_canon_sha256":"b7e09dd5d66bc57bd91281e02938e90215573937cc5769b4915af422d8454621"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:13:44.864185Z","signature_b64":"3TY1DTp2mn6y8r1ZF6BaccANwxS1Rn7VBXWhh7cZS4FzPAx+TCBTxCNUwf8ihD8lq6OzRjpoj2Q+zG2sMDq0Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1b3c60e24d49c6cb2dc3be43523f093a6ba54c54838dec733d4ba24e867e7264","last_reissued_at":"2026-05-18T00:13:44.863486Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:13:44.863486Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Exploring Hidden Dimensions in Parallelizing Convolutional Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC","cs.NE"],"primary_cat":"cs.LG","authors_text":"Alex Aiken, Charles R. Qi, Sina Lin, Zhihao Jia","submitted_at":"2018-02-14T02:00:40Z","abstract_excerpt":"The past few years have witnessed growth in the computational requirements for training deep convolutional neural networks. Current approaches parallelize training onto multiple devices by applying a single parallelization strategy (e.g., data or model parallelism) to all layers in a network. Although easy to reason about, these approaches result in suboptimal runtime performance in large-scale distributed training, since different layers in a network may prefer different parallelization strategies. In this paper, we propose layer-wise parallelism that allows each layer in a network to use an "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1802.04924","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"},"aliases":[{"alias_kind":"arxiv","alias_value":"1802.04924","created_at":"2026-05-18T00:13:44.863591+00:00"},{"alias_kind":"arxiv_version","alias_value":"1802.04924v2","created_at":"2026-05-18T00:13:44.863591+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1802.04924","created_at":"2026-05-18T00:13:44.863591+00:00"},{"alias_kind":"pith_short_12","alias_value":"DM6GBYSNJHDM","created_at":"2026-05-18T12:32:19.392346+00:00"},{"alias_kind":"pith_short_16","alias_value":"DM6GBYSNJHDMWLOD","created_at":"2026-05-18T12:32:19.392346+00:00"},{"alias_kind":"pith_short_8","alias_value":"DM6GBYSN","created_at":"2026-05-18T12:32:19.392346+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.02473","citing_title":"Analyzing Reverse Address Translation Overheads in Multi-GPU Scale-Up Pods","ref_index":52,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DM6GBYSNJHDMWLODXZBVEPYJHJ","json":"https://pith.science/pith/DM6GBYSNJHDMWLODXZBVEPYJHJ.json","graph_json":"https://pith.science/api/pith-number/DM6GBYSNJHDMWLODXZBVEPYJHJ/graph.json","events_json":"https://pith.science/api/pith-number/DM6GBYSNJHDMWLODXZBVEPYJHJ/events.json","paper":"https://pith.science/paper/DM6GBYSN"},"agent_actions":{"view_html":"https://pith.science/pith/DM6GBYSNJHDMWLODXZBVEPYJHJ","download_json":"https://pith.science/pith/DM6GBYSNJHDMWLODXZBVEPYJHJ.json","view_paper":"https://pith.science/paper/DM6GBYSN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1802.04924&json=true","fetch_graph":"https://pith.science/api/pith-number/DM6GBYSNJHDMWLODXZBVEPYJHJ/graph.json","fetch_events":"https://pith.science/api/pith-number/DM6GBYSNJHDMWLODXZBVEPYJHJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DM6GBYSNJHDMWLODXZBVEPYJHJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DM6GBYSNJHDMWLODXZBVEPYJHJ/action/storage_attestation","attest_author":"https://pith.science/pith/DM6GBYSNJHDMWLODXZBVEPYJHJ/action/author_attestation","sign_citation":"https://pith.science/pith/DM6GBYSNJHDMWLODXZBVEPYJHJ/action/citation_signature","submit_replication":"https://pith.science/pith/DM6GBYSNJHDMWLODXZBVEPYJHJ/action/replication_record"}},"created_at":"2026-05-18T00:13:44.863591+00:00","updated_at":"2026-05-18T00:13:44.863591+00:00"}