{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2013:CAHQ3M3GVY7LAZYET42V64QPMZ","short_pith_number":"pith:CAHQ3M3G","canonical_record":{"source":{"id":"1303.4664","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2013-03-19T17:00:22Z","cross_cats_sorted":[],"title_canon_sha256":"a199bb56a365695f679830d4be2692d1df3c1a34f03f414a4c28340c17bb54ee","abstract_canon_sha256":"50e87083227422708a0c286c19fb39454bd1f1a6c21225fef2a29f15277e4c2d"},"schema_version":"1.0"},"canonical_sha256":"100f0db366ae3eb067049f355f720f665f8fc5f297fa3583fa63db9e48d39119","source":{"kind":"arxiv","id":"1303.4664","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1303.4664","created_at":"2026-05-18T03:30:26Z"},{"alias_kind":"arxiv_version","alias_value":"1303.4664v1","created_at":"2026-05-18T03:30:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1303.4664","created_at":"2026-05-18T03:30:26Z"},{"alias_kind":"pith_short_12","alias_value":"CAHQ3M3GVY7L","created_at":"2026-05-18T12:27:40Z"},{"alias_kind":"pith_short_16","alias_value":"CAHQ3M3GVY7LAZYE","created_at":"2026-05-18T12:27:40Z"},{"alias_kind":"pith_short_8","alias_value":"CAHQ3M3G","created_at":"2026-05-18T12:27:40Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2013:CAHQ3M3GVY7LAZYET42V64QPMZ","target":"record","payload":{"canonical_record":{"source":{"id":"1303.4664","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2013-03-19T17:00:22Z","cross_cats_sorted":[],"title_canon_sha256":"a199bb56a365695f679830d4be2692d1df3c1a34f03f414a4c28340c17bb54ee","abstract_canon_sha256":"50e87083227422708a0c286c19fb39454bd1f1a6c21225fef2a29f15277e4c2d"},"schema_version":"1.0"},"canonical_sha256":"100f0db366ae3eb067049f355f720f665f8fc5f297fa3583fa63db9e48d39119","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:30:26.583048Z","signature_b64":"0OWbtvZjsWZ7ZYuYUTKxJEy80zlb36ZVSnEhLU7I95Vr7F42v3TPDsWe5yRWWB0vFELaFTr6TSZEBI0WuFeuCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"100f0db366ae3eb067049f355f720f665f8fc5f297fa3583fa63db9e48d39119","last_reissued_at":"2026-05-18T03:30:26.582497Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:30:26.582497Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1303.4664","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":"2026-05-18T03:30:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"L8QQEvTER186aLrEIe/WqAhsaVur40JXBVC0WAFkce6rPPkzB41n1JNIKfhPsenTMQqjP3aGBk3t9BtK90oGCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-05T03:55:32.898937Z"},"content_sha256":"b2279ab3aea32286cfbd539027d0ff4e8a8ca3869ad5d89784b42d9723e96ded","schema_version":"1.0","event_id":"sha256:b2279ab3aea32286cfbd539027d0ff4e8a8ca3869ad5d89784b42d9723e96ded"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2013:CAHQ3M3GVY7LAZYET42V64QPMZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Large-Scale Learning with Less RAM via Randomization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Daniel Golovin, D. Sculley, H. Brendan McMahan, Michael Young","submitted_at":"2013-03-19T17:00:22Z","abstract_excerpt":"We reduce the memory footprint of popular large-scale online learning methods by projecting our weight vector onto a coarse discrete set using randomized rounding. Compared to standard 32-bit float encodings, this reduces RAM usage by more than 50% during training and by up to 95% when making predictions from a fixed model, with almost no loss in accuracy. We also show that randomized counting can be used to implement per-coordinate learning rates, improving model quality with little additional RAM. We prove these memory-saving methods achieve regret guarantees similar to their exact variants."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1303.4664","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":"2026-05-18T03:30:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BHoLF6loB99bVHjhTT6GtQVzFJN4lWMzzpw3wMOqByHFVbGghe9LmlGyTN9qrSPGsWo+6JK/+gMO/bGL+p8dBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-05T03:55:32.899578Z"},"content_sha256":"319ee8be5498bbf4d65ab7d7c864db17508011109e94b13948349b0ceb79225b","schema_version":"1.0","event_id":"sha256:319ee8be5498bbf4d65ab7d7c864db17508011109e94b13948349b0ceb79225b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CAHQ3M3GVY7LAZYET42V64QPMZ/bundle.json","state_url":"https://pith.science/pith/CAHQ3M3GVY7LAZYET42V64QPMZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CAHQ3M3GVY7LAZYET42V64QPMZ/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":"2026-06-05T03:55:32Z","links":{"resolver":"https://pith.science/pith/CAHQ3M3GVY7LAZYET42V64QPMZ","bundle":"https://pith.science/pith/CAHQ3M3GVY7LAZYET42V64QPMZ/bundle.json","state":"https://pith.science/pith/CAHQ3M3GVY7LAZYET42V64QPMZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CAHQ3M3GVY7LAZYET42V64QPMZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2013:CAHQ3M3GVY7LAZYET42V64QPMZ","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":"50e87083227422708a0c286c19fb39454bd1f1a6c21225fef2a29f15277e4c2d","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2013-03-19T17:00:22Z","title_canon_sha256":"a199bb56a365695f679830d4be2692d1df3c1a34f03f414a4c28340c17bb54ee"},"schema_version":"1.0","source":{"id":"1303.4664","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1303.4664","created_at":"2026-05-18T03:30:26Z"},{"alias_kind":"arxiv_version","alias_value":"1303.4664v1","created_at":"2026-05-18T03:30:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1303.4664","created_at":"2026-05-18T03:30:26Z"},{"alias_kind":"pith_short_12","alias_value":"CAHQ3M3GVY7L","created_at":"2026-05-18T12:27:40Z"},{"alias_kind":"pith_short_16","alias_value":"CAHQ3M3GVY7LAZYE","created_at":"2026-05-18T12:27:40Z"},{"alias_kind":"pith_short_8","alias_value":"CAHQ3M3G","created_at":"2026-05-18T12:27:40Z"}],"graph_snapshots":[{"event_id":"sha256:319ee8be5498bbf4d65ab7d7c864db17508011109e94b13948349b0ceb79225b","target":"graph","created_at":"2026-05-18T03:30:26Z","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":"We reduce the memory footprint of popular large-scale online learning methods by projecting our weight vector onto a coarse discrete set using randomized rounding. Compared to standard 32-bit float encodings, this reduces RAM usage by more than 50% during training and by up to 95% when making predictions from a fixed model, with almost no loss in accuracy. We also show that randomized counting can be used to implement per-coordinate learning rates, improving model quality with little additional RAM. We prove these memory-saving methods achieve regret guarantees similar to their exact variants.","authors_text":"Daniel Golovin, D. Sculley, H. Brendan McMahan, Michael Young","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2013-03-19T17:00:22Z","title":"Large-Scale Learning with Less RAM via Randomization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1303.4664","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:b2279ab3aea32286cfbd539027d0ff4e8a8ca3869ad5d89784b42d9723e96ded","target":"record","created_at":"2026-05-18T03:30:26Z","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":"50e87083227422708a0c286c19fb39454bd1f1a6c21225fef2a29f15277e4c2d","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2013-03-19T17:00:22Z","title_canon_sha256":"a199bb56a365695f679830d4be2692d1df3c1a34f03f414a4c28340c17bb54ee"},"schema_version":"1.0","source":{"id":"1303.4664","kind":"arxiv","version":1}},"canonical_sha256":"100f0db366ae3eb067049f355f720f665f8fc5f297fa3583fa63db9e48d39119","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"100f0db366ae3eb067049f355f720f665f8fc5f297fa3583fa63db9e48d39119","first_computed_at":"2026-05-18T03:30:26.582497Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T03:30:26.582497Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"0OWbtvZjsWZ7ZYuYUTKxJEy80zlb36ZVSnEhLU7I95Vr7F42v3TPDsWe5yRWWB0vFELaFTr6TSZEBI0WuFeuCQ==","signature_status":"signed_v1","signed_at":"2026-05-18T03:30:26.583048Z","signed_message":"canonical_sha256_bytes"},"source_id":"1303.4664","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b2279ab3aea32286cfbd539027d0ff4e8a8ca3869ad5d89784b42d9723e96ded","sha256:319ee8be5498bbf4d65ab7d7c864db17508011109e94b13948349b0ceb79225b"],"state_sha256":"15ca855655ed8cdf42e2f23888205483283c0ffea328bf96d6a337823ddd381a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mKEvafecBK6WSWQkfse+fJkg/2i012zekskoGcNlF2lHwF9/kAvBjAtot1YYE9LrVW805QJjh2KXKe1uT+sqBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-05T03:55:32.903299Z","bundle_sha256":"fd2ad6d0c15fd4943c47afb1c75182b6d1292dd6f938769c90abfd635e941d2f"}}