{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:ZPTEBHIMA5BTCFDZEPIM3AC3TI","short_pith_number":"pith:ZPTEBHIM","canonical_record":{"source":{"id":"1704.06363","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2017-04-20T23:45:27Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"838fa2582fa7252b7858f41f1f0b1939d874fe3d520d3865b77e332ea15aff20","abstract_canon_sha256":"73f025ec92d8e3dde3f57c3f87cc3beb2bbef928000b29867077383e6ad9dec4"},"schema_version":"1.0"},"canonical_sha256":"cbe6409d0c074331147923d0cd805b9a343831ebc131fde517117c166d1bc72f","source":{"kind":"arxiv","id":"1704.06363","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1704.06363","created_at":"2026-05-18T00:45:59Z"},{"alias_kind":"arxiv_version","alias_value":"1704.06363v1","created_at":"2026-05-18T00:45:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1704.06363","created_at":"2026-05-18T00:45:59Z"},{"alias_kind":"pith_short_12","alias_value":"ZPTEBHIMA5BT","created_at":"2026-05-18T12:31:59Z"},{"alias_kind":"pith_short_16","alias_value":"ZPTEBHIMA5BTCFDZ","created_at":"2026-05-18T12:31:59Z"},{"alias_kind":"pith_short_8","alias_value":"ZPTEBHIM","created_at":"2026-05-18T12:31:59Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:ZPTEBHIMA5BTCFDZEPIM3AC3TI","target":"record","payload":{"canonical_record":{"source":{"id":"1704.06363","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2017-04-20T23:45:27Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"838fa2582fa7252b7858f41f1f0b1939d874fe3d520d3865b77e332ea15aff20","abstract_canon_sha256":"73f025ec92d8e3dde3f57c3f87cc3beb2bbef928000b29867077383e6ad9dec4"},"schema_version":"1.0"},"canonical_sha256":"cbe6409d0c074331147923d0cd805b9a343831ebc131fde517117c166d1bc72f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:45:59.895733Z","signature_b64":"as13yxgzjlfRZmjOYoIiwiUjtaDHlJK3xJFrG2R2YF9G7F1QJrKYtnQS2vpZQelZH+U5FreFQJp0A3sTonnrDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cbe6409d0c074331147923d0cd805b9a343831ebc131fde517117c166d1bc72f","last_reissued_at":"2026-05-18T00:45:59.895074Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:45:59.895074Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1704.06363","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":"2026-05-18T00:45:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WcfF3ZkuBtLIdefwVBt2ibkoI0pMMu1FoBPJ8V77yAO8qdRrhh+z1MDyAnO7gIyc4MPaGu3op5NcoWQPrmTyDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-22T04:25:06.162096Z"},"content_sha256":"74b39aca38628f6c8360b41a17b3d406bc62a3041044d6ab0544f9938d75e4f2","schema_version":"1.0","event_id":"sha256:74b39aca38628f6c8360b41a17b3d406bc62a3041044d6ab0544f9938d75e4f2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:ZPTEBHIMA5BTCFDZEPIM3AC3TI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Hard Mixtures of Experts for Large Scale Weakly Supervised Vision","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.CV","authors_text":"Arthur Szlam, Marc'Aurelio Ranzato, Sam Gross","submitted_at":"2017-04-20T23:45:27Z","abstract_excerpt":"Training convolutional networks (CNN's) that fit on a single GPU with minibatch stochastic gradient descent has become effective in practice. However, there is still no effective method for training large CNN's that do not fit in the memory of a few GPU cards, or for parallelizing CNN training. In this work we show that a simple hard mixture of experts model can be efficiently trained to good effect on large scale hashtag (multilabel) prediction tasks. Mixture of experts models are not new (Jacobs et. al. 1991, Collobert et. al. 2003), but in the past, researchers have had to devise sophistica"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1704.06363","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":"2026-05-18T00:45:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kK/RKuNnw3hg7DJyTUYjBZtozsCxHy+SMsWkGLZuTN+cqBtcpjtHPnunsYkG5rEZbT8m+5Vd84jRfgaD4CZODQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-22T04:25:06.162795Z"},"content_sha256":"96d4d89faa3215fd0f0710c16b72855c3887c689a9d5014c55b96d4f357d618b","schema_version":"1.0","event_id":"sha256:96d4d89faa3215fd0f0710c16b72855c3887c689a9d5014c55b96d4f357d618b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZPTEBHIMA5BTCFDZEPIM3AC3TI/bundle.json","state_url":"https://pith.science/pith/ZPTEBHIMA5BTCFDZEPIM3AC3TI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZPTEBHIMA5BTCFDZEPIM3AC3TI/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":"2026-05-22T04:25:06Z","links":{"resolver":"https://pith.science/pith/ZPTEBHIMA5BTCFDZEPIM3AC3TI","bundle":"https://pith.science/pith/ZPTEBHIMA5BTCFDZEPIM3AC3TI/bundle.json","state":"https://pith.science/pith/ZPTEBHIMA5BTCFDZEPIM3AC3TI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZPTEBHIMA5BTCFDZEPIM3AC3TI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:ZPTEBHIMA5BTCFDZEPIM3AC3TI","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":"73f025ec92d8e3dde3f57c3f87cc3beb2bbef928000b29867077383e6ad9dec4","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2017-04-20T23:45:27Z","title_canon_sha256":"838fa2582fa7252b7858f41f1f0b1939d874fe3d520d3865b77e332ea15aff20"},"schema_version":"1.0","source":{"id":"1704.06363","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1704.06363","created_at":"2026-05-18T00:45:59Z"},{"alias_kind":"arxiv_version","alias_value":"1704.06363v1","created_at":"2026-05-18T00:45:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1704.06363","created_at":"2026-05-18T00:45:59Z"},{"alias_kind":"pith_short_12","alias_value":"ZPTEBHIMA5BT","created_at":"2026-05-18T12:31:59Z"},{"alias_kind":"pith_short_16","alias_value":"ZPTEBHIMA5BTCFDZ","created_at":"2026-05-18T12:31:59Z"},{"alias_kind":"pith_short_8","alias_value":"ZPTEBHIM","created_at":"2026-05-18T12:31:59Z"}],"graph_snapshots":[{"event_id":"sha256:96d4d89faa3215fd0f0710c16b72855c3887c689a9d5014c55b96d4f357d618b","target":"graph","created_at":"2026-05-18T00:45:59Z","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":"Training convolutional networks (CNN's) that fit on a single GPU with minibatch stochastic gradient descent has become effective in practice. However, there is still no effective method for training large CNN's that do not fit in the memory of a few GPU cards, or for parallelizing CNN training. In this work we show that a simple hard mixture of experts model can be efficiently trained to good effect on large scale hashtag (multilabel) prediction tasks. Mixture of experts models are not new (Jacobs et. al. 1991, Collobert et. al. 2003), but in the past, researchers have had to devise sophistica","authors_text":"Arthur Szlam, Marc'Aurelio Ranzato, Sam Gross","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2017-04-20T23:45:27Z","title":"Hard Mixtures of Experts for Large Scale Weakly Supervised Vision"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1704.06363","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":"sha256:74b39aca38628f6c8360b41a17b3d406bc62a3041044d6ab0544f9938d75e4f2","target":"record","created_at":"2026-05-18T00:45:59Z","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":"73f025ec92d8e3dde3f57c3f87cc3beb2bbef928000b29867077383e6ad9dec4","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2017-04-20T23:45:27Z","title_canon_sha256":"838fa2582fa7252b7858f41f1f0b1939d874fe3d520d3865b77e332ea15aff20"},"schema_version":"1.0","source":{"id":"1704.06363","kind":"arxiv","version":1}},"canonical_sha256":"cbe6409d0c074331147923d0cd805b9a343831ebc131fde517117c166d1bc72f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cbe6409d0c074331147923d0cd805b9a343831ebc131fde517117c166d1bc72f","first_computed_at":"2026-05-18T00:45:59.895074Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:45:59.895074Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"as13yxgzjlfRZmjOYoIiwiUjtaDHlJK3xJFrG2R2YF9G7F1QJrKYtnQS2vpZQelZH+U5FreFQJp0A3sTonnrDQ==","signature_status":"signed_v1","signed_at":"2026-05-18T00:45:59.895733Z","signed_message":"canonical_sha256_bytes"},"source_id":"1704.06363","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:74b39aca38628f6c8360b41a17b3d406bc62a3041044d6ab0544f9938d75e4f2","sha256:96d4d89faa3215fd0f0710c16b72855c3887c689a9d5014c55b96d4f357d618b"],"state_sha256":"5404eb3b30edb20415c8797b2691ecc5c67bebaa791375bdca2579d556eddff7"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gKQhp/a25j6fZtAUu8UBy30u6y+WoMoq5rmalfVR6PuwjnS2+cvZDFq6pSi56udQMX4W6HsJa2ukpL490D2eBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-22T04:25:06.166976Z","bundle_sha256":"a1376d601608979d9f7b9495755917ce044b91c74b731c74f9d29f984bff5dc5"}}