{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:E7RDIIRRWGU6NY7ESZEPJ5UV2W","short_pith_number":"pith:E7RDIIRR","canonical_record":{"source":{"id":"1703.05880","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-03-17T03:38:48Z","cross_cats_sorted":["cs.LG","cs.SD","eess.AS"],"title_canon_sha256":"8418987141025d8d7026190e9b690bc0a4811e7c986eb9b594eeb311976e0b75","abstract_canon_sha256":"d74a9dc8eb7e68593e970a1c7e802f0b2e216f07b24deefab2f2959af789e098"},"schema_version":"1.0"},"canonical_sha256":"27e2342231b1a9e6e3e49648f4f695d5a04b79c5245c4c60f2b4a6bec88fc7b6","source":{"kind":"arxiv","id":"1703.05880","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1703.05880","created_at":"2026-05-17T23:59:03Z"},{"alias_kind":"arxiv_version","alias_value":"1703.05880v2","created_at":"2026-05-17T23:59:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1703.05880","created_at":"2026-05-17T23:59:03Z"},{"alias_kind":"pith_short_12","alias_value":"E7RDIIRRWGU6","created_at":"2026-05-18T12:31:12Z"},{"alias_kind":"pith_short_16","alias_value":"E7RDIIRRWGU6NY7E","created_at":"2026-05-18T12:31:12Z"},{"alias_kind":"pith_short_8","alias_value":"E7RDIIRR","created_at":"2026-05-18T12:31:12Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:E7RDIIRRWGU6NY7ESZEPJ5UV2W","target":"record","payload":{"canonical_record":{"source":{"id":"1703.05880","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-03-17T03:38:48Z","cross_cats_sorted":["cs.LG","cs.SD","eess.AS"],"title_canon_sha256":"8418987141025d8d7026190e9b690bc0a4811e7c986eb9b594eeb311976e0b75","abstract_canon_sha256":"d74a9dc8eb7e68593e970a1c7e802f0b2e216f07b24deefab2f2959af789e098"},"schema_version":"1.0"},"canonical_sha256":"27e2342231b1a9e6e3e49648f4f695d5a04b79c5245c4c60f2b4a6bec88fc7b6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:59:03.794901Z","signature_b64":"ewBowip+MSVRA4CAug0BDZA20i5Id/wQhfYfZZRHL6ZR5cj86W8moc1nMtadXSm/mvBwCI5Bo8WHxKG28GmjBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"27e2342231b1a9e6e3e49648f4f695d5a04b79c5245c4c60f2b4a6bec88fc7b6","last_reissued_at":"2026-05-17T23:59:03.794350Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:59:03.794350Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1703.05880","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":"2026-05-17T23:59:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tXUan1YTSsEGgAxrUmhpzHP3O81i/LhlA5wJYOp+BUqy93n/7UceIWZWBZQfoeWMlIQaHCg0UWZPsmzUVzRwDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T05:19:28.066640Z"},"content_sha256":"33e0968aaaa8410e0bebefbaed0111b7a78f6b1eca654c4837e4c06e7355677d","schema_version":"1.0","event_id":"sha256:33e0968aaaa8410e0bebefbaed0111b7a78f6b1eca654c4837e4c06e7355677d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:E7RDIIRRWGU6NY7ESZEPJ5UV2W","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Empirical Evaluation of Parallel Training Algorithms on Acoustic Modeling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Binbin Zhang, Dong Yu, Lei Xie, Wenpeng Li","submitted_at":"2017-03-17T03:38:48Z","abstract_excerpt":"Deep learning models (DLMs) are state-of-the-art techniques in speech recognition. However, training good DLMs can be time consuming especially for production-size models and corpora. Although several parallel training algorithms have been proposed to improve training efficiency, there is no clear guidance on which one to choose for the task in hand due to lack of systematic and fair comparison among them. In this paper we aim at filling this gap by comparing four popular parallel training algorithms in speech recognition, namely asynchronous stochastic gradient descent (ASGD), blockwise model"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1703.05880","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":"2026-05-17T23:59:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iqr12etD6FnX1TuADH/Z4GVlvaKZdL2bz+QF3ExxW6ymIptFxBQqNwlpPqrgkhEtJq4hich12wUNArJR4EV9Aw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T05:19:28.067265Z"},"content_sha256":"e8d175a7bc4d700e0b99b450823983a1d57a6a1ce32734d1296694e3baa5b36d","schema_version":"1.0","event_id":"sha256:e8d175a7bc4d700e0b99b450823983a1d57a6a1ce32734d1296694e3baa5b36d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/E7RDIIRRWGU6NY7ESZEPJ5UV2W/bundle.json","state_url":"https://pith.science/pith/E7RDIIRRWGU6NY7ESZEPJ5UV2W/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/E7RDIIRRWGU6NY7ESZEPJ5UV2W/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":"2026-05-27T05:19:28Z","links":{"resolver":"https://pith.science/pith/E7RDIIRRWGU6NY7ESZEPJ5UV2W","bundle":"https://pith.science/pith/E7RDIIRRWGU6NY7ESZEPJ5UV2W/bundle.json","state":"https://pith.science/pith/E7RDIIRRWGU6NY7ESZEPJ5UV2W/state.json","well_known_bundle":"https://pith.science/.well-known/pith/E7RDIIRRWGU6NY7ESZEPJ5UV2W/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:E7RDIIRRWGU6NY7ESZEPJ5UV2W","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":"d74a9dc8eb7e68593e970a1c7e802f0b2e216f07b24deefab2f2959af789e098","cross_cats_sorted":["cs.LG","cs.SD","eess.AS"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-03-17T03:38:48Z","title_canon_sha256":"8418987141025d8d7026190e9b690bc0a4811e7c986eb9b594eeb311976e0b75"},"schema_version":"1.0","source":{"id":"1703.05880","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1703.05880","created_at":"2026-05-17T23:59:03Z"},{"alias_kind":"arxiv_version","alias_value":"1703.05880v2","created_at":"2026-05-17T23:59:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1703.05880","created_at":"2026-05-17T23:59:03Z"},{"alias_kind":"pith_short_12","alias_value":"E7RDIIRRWGU6","created_at":"2026-05-18T12:31:12Z"},{"alias_kind":"pith_short_16","alias_value":"E7RDIIRRWGU6NY7E","created_at":"2026-05-18T12:31:12Z"},{"alias_kind":"pith_short_8","alias_value":"E7RDIIRR","created_at":"2026-05-18T12:31:12Z"}],"graph_snapshots":[{"event_id":"sha256:e8d175a7bc4d700e0b99b450823983a1d57a6a1ce32734d1296694e3baa5b36d","target":"graph","created_at":"2026-05-17T23:59:03Z","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":"Deep learning models (DLMs) are state-of-the-art techniques in speech recognition. However, training good DLMs can be time consuming especially for production-size models and corpora. Although several parallel training algorithms have been proposed to improve training efficiency, there is no clear guidance on which one to choose for the task in hand due to lack of systematic and fair comparison among them. In this paper we aim at filling this gap by comparing four popular parallel training algorithms in speech recognition, namely asynchronous stochastic gradient descent (ASGD), blockwise model","authors_text":"Binbin Zhang, Dong Yu, Lei Xie, Wenpeng Li","cross_cats":["cs.LG","cs.SD","eess.AS"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-03-17T03:38:48Z","title":"Empirical Evaluation of Parallel Training Algorithms on Acoustic Modeling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1703.05880","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":"sha256:33e0968aaaa8410e0bebefbaed0111b7a78f6b1eca654c4837e4c06e7355677d","target":"record","created_at":"2026-05-17T23:59:03Z","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":"d74a9dc8eb7e68593e970a1c7e802f0b2e216f07b24deefab2f2959af789e098","cross_cats_sorted":["cs.LG","cs.SD","eess.AS"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-03-17T03:38:48Z","title_canon_sha256":"8418987141025d8d7026190e9b690bc0a4811e7c986eb9b594eeb311976e0b75"},"schema_version":"1.0","source":{"id":"1703.05880","kind":"arxiv","version":2}},"canonical_sha256":"27e2342231b1a9e6e3e49648f4f695d5a04b79c5245c4c60f2b4a6bec88fc7b6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"27e2342231b1a9e6e3e49648f4f695d5a04b79c5245c4c60f2b4a6bec88fc7b6","first_computed_at":"2026-05-17T23:59:03.794350Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:59:03.794350Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ewBowip+MSVRA4CAug0BDZA20i5Id/wQhfYfZZRHL6ZR5cj86W8moc1nMtadXSm/mvBwCI5Bo8WHxKG28GmjBg==","signature_status":"signed_v1","signed_at":"2026-05-17T23:59:03.794901Z","signed_message":"canonical_sha256_bytes"},"source_id":"1703.05880","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:33e0968aaaa8410e0bebefbaed0111b7a78f6b1eca654c4837e4c06e7355677d","sha256:e8d175a7bc4d700e0b99b450823983a1d57a6a1ce32734d1296694e3baa5b36d"],"state_sha256":"3abe01cd287d6e67c0210b3fbcdbe122fb51820b8d71f1b8b1da272452d2d21e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fuRST4xVLVfYKz2PQM4K7w8h8EiuwVdfDpeuUtdtvvVonlSb5IeS5ppnMURBOttsZmXYUP4Rmdm20qrlUEsXAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T05:19:28.071214Z","bundle_sha256":"062bce60cb46efe3069cb4019c679492157e576dcff9c09b111eebfdb6ddecca"}}