{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2013:XTJMNCEUKKELS43BMUFHD26E3H","short_pith_number":"pith:XTJMNCEU","canonical_record":{"source":{"id":"1301.3641","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2013-01-16T10:10:23Z","cross_cats_sorted":["cs.NE","stat.ML"],"title_canon_sha256":"e29fc5029e389b573a48fd4918db9734b5c35d64ad6fd45d157e481ce1b4f1c1","abstract_canon_sha256":"159ab6e1453ab298d5eab10afb681e7b91f52c2a88a04d57b18c9d7640fff2af"},"schema_version":"1.0"},"canonical_sha256":"bcd2c688945288b97361650a71ebc4d9d3be8514624236834d4620b2916322e5","source":{"kind":"arxiv","id":"1301.3641","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1301.3641","created_at":"2026-05-18T03:26:42Z"},{"alias_kind":"arxiv_version","alias_value":"1301.3641v3","created_at":"2026-05-18T03:26:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1301.3641","created_at":"2026-05-18T03:26:42Z"},{"alias_kind":"pith_short_12","alias_value":"XTJMNCEUKKEL","created_at":"2026-05-18T12:28:06Z"},{"alias_kind":"pith_short_16","alias_value":"XTJMNCEUKKELS43B","created_at":"2026-05-18T12:28:06Z"},{"alias_kind":"pith_short_8","alias_value":"XTJMNCEU","created_at":"2026-05-18T12:28:06Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2013:XTJMNCEUKKELS43BMUFHD26E3H","target":"record","payload":{"canonical_record":{"source":{"id":"1301.3641","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2013-01-16T10:10:23Z","cross_cats_sorted":["cs.NE","stat.ML"],"title_canon_sha256":"e29fc5029e389b573a48fd4918db9734b5c35d64ad6fd45d157e481ce1b4f1c1","abstract_canon_sha256":"159ab6e1453ab298d5eab10afb681e7b91f52c2a88a04d57b18c9d7640fff2af"},"schema_version":"1.0"},"canonical_sha256":"bcd2c688945288b97361650a71ebc4d9d3be8514624236834d4620b2916322e5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:26:42.595251Z","signature_b64":"y+GO+wNngi5mkgFKgulO7Tb/f+eCvxhCqhQk+x+bZEybUhWpddw+MgLAIUqsoySLuyyhH7w5XGiQoApO1k4KDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bcd2c688945288b97361650a71ebc4d9d3be8514624236834d4620b2916322e5","last_reissued_at":"2026-05-18T03:26:42.594723Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:26:42.594723Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1301.3641","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T03:26:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZF+kT+YiVE9Ul68RPD4SOBQAEqIPEPvohqvyWtHJjTljRjINVMZz3S2hP0hKe2cZMjSWpTB3OwvzL7AhKhs3BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-07T05:00:31.376355Z"},"content_sha256":"10dbd3ffb1adb782a5f44a709f2e575e88b01754ce3b809014343aac76d10a6a","schema_version":"1.0","event_id":"sha256:10dbd3ffb1adb782a5f44a709f2e575e88b01754ce3b809014343aac76d10a6a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2013:XTJMNCEUKKELS43BMUFHD26E3H","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Training Neural Networks with Stochastic Hessian-Free Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NE","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ryan Kiros","submitted_at":"2013-01-16T10:10:23Z","abstract_excerpt":"Hessian-free (HF) optimization has been successfully used for training deep autoencoders and recurrent networks. HF uses the conjugate gradient algorithm to construct update directions through curvature-vector products that can be computed on the same order of time as gradients. In this paper we exploit this property and study stochastic HF with gradient and curvature mini-batches independent of the dataset size. We modify Martens' HF for these settings and integrate dropout, a method for preventing co-adaptation of feature detectors, to guard against overfitting. Stochastic Hessian-free optim"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1301.3641","kind":"arxiv","version":3},"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":"2026-05-18T03:26:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kSveGYHx+oMTW2/v6rj98jvGmCY/pV9Wu7RoG6jIkvdrgO85K9EjjafG/4mVXSiFJViam7XWhRnUD4DFyNSACA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-07T05:00:31.376944Z"},"content_sha256":"8c7215110a33d676f7fee9e7951c89a8a6860be7f8a0d09baf6f9af0f4319e8c","schema_version":"1.0","event_id":"sha256:8c7215110a33d676f7fee9e7951c89a8a6860be7f8a0d09baf6f9af0f4319e8c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XTJMNCEUKKELS43BMUFHD26E3H/bundle.json","state_url":"https://pith.science/pith/XTJMNCEUKKELS43BMUFHD26E3H/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XTJMNCEUKKELS43BMUFHD26E3H/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":"2026-06-07T05:00:31Z","links":{"resolver":"https://pith.science/pith/XTJMNCEUKKELS43BMUFHD26E3H","bundle":"https://pith.science/pith/XTJMNCEUKKELS43BMUFHD26E3H/bundle.json","state":"https://pith.science/pith/XTJMNCEUKKELS43BMUFHD26E3H/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XTJMNCEUKKELS43BMUFHD26E3H/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2013:XTJMNCEUKKELS43BMUFHD26E3H","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":"159ab6e1453ab298d5eab10afb681e7b91f52c2a88a04d57b18c9d7640fff2af","cross_cats_sorted":["cs.NE","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2013-01-16T10:10:23Z","title_canon_sha256":"e29fc5029e389b573a48fd4918db9734b5c35d64ad6fd45d157e481ce1b4f1c1"},"schema_version":"1.0","source":{"id":"1301.3641","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1301.3641","created_at":"2026-05-18T03:26:42Z"},{"alias_kind":"arxiv_version","alias_value":"1301.3641v3","created_at":"2026-05-18T03:26:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1301.3641","created_at":"2026-05-18T03:26:42Z"},{"alias_kind":"pith_short_12","alias_value":"XTJMNCEUKKEL","created_at":"2026-05-18T12:28:06Z"},{"alias_kind":"pith_short_16","alias_value":"XTJMNCEUKKELS43B","created_at":"2026-05-18T12:28:06Z"},{"alias_kind":"pith_short_8","alias_value":"XTJMNCEU","created_at":"2026-05-18T12:28:06Z"}],"graph_snapshots":[{"event_id":"sha256:8c7215110a33d676f7fee9e7951c89a8a6860be7f8a0d09baf6f9af0f4319e8c","target":"graph","created_at":"2026-05-18T03:26:42Z","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":"Hessian-free (HF) optimization has been successfully used for training deep autoencoders and recurrent networks. HF uses the conjugate gradient algorithm to construct update directions through curvature-vector products that can be computed on the same order of time as gradients. In this paper we exploit this property and study stochastic HF with gradient and curvature mini-batches independent of the dataset size. We modify Martens' HF for these settings and integrate dropout, a method for preventing co-adaptation of feature detectors, to guard against overfitting. Stochastic Hessian-free optim","authors_text":"Ryan Kiros","cross_cats":["cs.NE","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2013-01-16T10:10:23Z","title":"Training Neural Networks with Stochastic Hessian-Free Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1301.3641","kind":"arxiv","version":3},"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":"sha256:10dbd3ffb1adb782a5f44a709f2e575e88b01754ce3b809014343aac76d10a6a","target":"record","created_at":"2026-05-18T03:26:42Z","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":"159ab6e1453ab298d5eab10afb681e7b91f52c2a88a04d57b18c9d7640fff2af","cross_cats_sorted":["cs.NE","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2013-01-16T10:10:23Z","title_canon_sha256":"e29fc5029e389b573a48fd4918db9734b5c35d64ad6fd45d157e481ce1b4f1c1"},"schema_version":"1.0","source":{"id":"1301.3641","kind":"arxiv","version":3}},"canonical_sha256":"bcd2c688945288b97361650a71ebc4d9d3be8514624236834d4620b2916322e5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bcd2c688945288b97361650a71ebc4d9d3be8514624236834d4620b2916322e5","first_computed_at":"2026-05-18T03:26:42.594723Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T03:26:42.594723Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"y+GO+wNngi5mkgFKgulO7Tb/f+eCvxhCqhQk+x+bZEybUhWpddw+MgLAIUqsoySLuyyhH7w5XGiQoApO1k4KDQ==","signature_status":"signed_v1","signed_at":"2026-05-18T03:26:42.595251Z","signed_message":"canonical_sha256_bytes"},"source_id":"1301.3641","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:10dbd3ffb1adb782a5f44a709f2e575e88b01754ce3b809014343aac76d10a6a","sha256:8c7215110a33d676f7fee9e7951c89a8a6860be7f8a0d09baf6f9af0f4319e8c"],"state_sha256":"28e799585c92de5f8db5f32c11869325d48382ac7c7b9b2ed9b906b202b35f9b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fxA7ztV5LzDWikONDYRW4za4jgOexsOrQTXIALuxEPEf2e3/Twy4sxmoS0NATnOTChwlD46FwDsWHcFUuiSQBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-07T05:00:31.382778Z","bundle_sha256":"b404b3deae47c853367d4aff9d9e15f55c3ee807c5425205e0bdac66a75acf92"}}