{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:763VWTWGN2PWALX4DYLEAP4VGM","short_pith_number":"pith:763VWTWG","canonical_record":{"source":{"id":"1810.03798","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-10-09T03:37:08Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"80055246e3acb0b3c223b9eea6b03117215f753c69bcfc29d13287b394f42a96","abstract_canon_sha256":"7ea83a9c95c097dc1bf4d5c6f2fb67c234f5cacc2817543d76ff3aaa14e3cc04"},"schema_version":"1.0"},"canonical_sha256":"ffb75b4ec66e9f602efc1e16403f953339db20dd7a64c5655c28f348f4da4cec","source":{"kind":"arxiv","id":"1810.03798","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1810.03798","created_at":"2026-05-18T00:03:44Z"},{"alias_kind":"arxiv_version","alias_value":"1810.03798v1","created_at":"2026-05-18T00:03:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1810.03798","created_at":"2026-05-18T00:03:44Z"},{"alias_kind":"pith_short_12","alias_value":"763VWTWGN2PW","created_at":"2026-05-18T12:32:11Z"},{"alias_kind":"pith_short_16","alias_value":"763VWTWGN2PWALX4","created_at":"2026-05-18T12:32:11Z"},{"alias_kind":"pith_short_8","alias_value":"763VWTWG","created_at":"2026-05-18T12:32:11Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:763VWTWGN2PWALX4DYLEAP4VGM","target":"record","payload":{"canonical_record":{"source":{"id":"1810.03798","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-10-09T03:37:08Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"80055246e3acb0b3c223b9eea6b03117215f753c69bcfc29d13287b394f42a96","abstract_canon_sha256":"7ea83a9c95c097dc1bf4d5c6f2fb67c234f5cacc2817543d76ff3aaa14e3cc04"},"schema_version":"1.0"},"canonical_sha256":"ffb75b4ec66e9f602efc1e16403f953339db20dd7a64c5655c28f348f4da4cec","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:03:44.492581Z","signature_b64":"tU1ieQ6Iy1zmkzfnS05tOK+n2wr1dzQ+MGeZ7wQZTNqffhB8nAv5Nd6R7KteYG4LMuStUzxKJGsl2av3ssYHAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ffb75b4ec66e9f602efc1e16403f953339db20dd7a64c5655c28f348f4da4cec","last_reissued_at":"2026-05-18T00:03:44.492143Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:03:44.492143Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1810.03798","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":"2026-05-18T00:03:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"K3vaX2+Tywc8fq5C2/wt4oMXbo7PzYPG07N1MFmw7V5H/+yIgkmhCQUVGMNhe+FY6ZgxERsGT5LcOz6wDKY1BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-28T22:09:56.664092Z"},"content_sha256":"0e9ac183842749a71bfbe21afd15edd62498446ee4d8243dd67bf6837a79616b","schema_version":"1.0","event_id":"sha256:0e9ac183842749a71bfbe21afd15edd62498446ee4d8243dd67bf6837a79616b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:763VWTWGN2PWALX4DYLEAP4VGM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"The Outer Product Structure of Neural Network Derivatives","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Craig Bakker, Michael J. Henry, Nathan O. Hodas","submitted_at":"2018-10-09T03:37:08Z","abstract_excerpt":"In this paper, we show that feedforward and recurrent neural networks exhibit an outer product derivative structure but that convolutional neural networks do not. This structure makes it possible to use higher-order information without needing approximations or infeasibly large amounts of memory, and it may also provide insights into the geometry of neural network optima. The ability to easily access these derivatives also suggests a new, geometric approach to regularization. We then discuss how this structure could be used to improve training methods, increase network robustness and generaliz"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1810.03798","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":"2026-05-18T00:03:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Nx8Tz/TTI/fJnpPMZFfz1xWWR4/BaZUILP6cImSzLvM2IJVgS2/Mnt9ecSJ2LvvtCWUE+6fthe0Oy5Nc+vjmAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-28T22:09:56.664761Z"},"content_sha256":"8983e83bba4d784dba010f41b32a9cf5e78e5dc6e7030b9c92152ce38b30625b","schema_version":"1.0","event_id":"sha256:8983e83bba4d784dba010f41b32a9cf5e78e5dc6e7030b9c92152ce38b30625b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/763VWTWGN2PWALX4DYLEAP4VGM/bundle.json","state_url":"https://pith.science/pith/763VWTWGN2PWALX4DYLEAP4VGM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/763VWTWGN2PWALX4DYLEAP4VGM/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":"2026-05-28T22:09:56Z","links":{"resolver":"https://pith.science/pith/763VWTWGN2PWALX4DYLEAP4VGM","bundle":"https://pith.science/pith/763VWTWGN2PWALX4DYLEAP4VGM/bundle.json","state":"https://pith.science/pith/763VWTWGN2PWALX4DYLEAP4VGM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/763VWTWGN2PWALX4DYLEAP4VGM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:763VWTWGN2PWALX4DYLEAP4VGM","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":"7ea83a9c95c097dc1bf4d5c6f2fb67c234f5cacc2817543d76ff3aaa14e3cc04","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-10-09T03:37:08Z","title_canon_sha256":"80055246e3acb0b3c223b9eea6b03117215f753c69bcfc29d13287b394f42a96"},"schema_version":"1.0","source":{"id":"1810.03798","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1810.03798","created_at":"2026-05-18T00:03:44Z"},{"alias_kind":"arxiv_version","alias_value":"1810.03798v1","created_at":"2026-05-18T00:03:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1810.03798","created_at":"2026-05-18T00:03:44Z"},{"alias_kind":"pith_short_12","alias_value":"763VWTWGN2PW","created_at":"2026-05-18T12:32:11Z"},{"alias_kind":"pith_short_16","alias_value":"763VWTWGN2PWALX4","created_at":"2026-05-18T12:32:11Z"},{"alias_kind":"pith_short_8","alias_value":"763VWTWG","created_at":"2026-05-18T12:32:11Z"}],"graph_snapshots":[{"event_id":"sha256:8983e83bba4d784dba010f41b32a9cf5e78e5dc6e7030b9c92152ce38b30625b","target":"graph","created_at":"2026-05-18T00:03:44Z","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":"In this paper, we show that feedforward and recurrent neural networks exhibit an outer product derivative structure but that convolutional neural networks do not. This structure makes it possible to use higher-order information without needing approximations or infeasibly large amounts of memory, and it may also provide insights into the geometry of neural network optima. The ability to easily access these derivatives also suggests a new, geometric approach to regularization. We then discuss how this structure could be used to improve training methods, increase network robustness and generaliz","authors_text":"Craig Bakker, Michael J. Henry, Nathan O. Hodas","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-10-09T03:37:08Z","title":"The Outer Product Structure of Neural Network Derivatives"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1810.03798","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":"sha256:0e9ac183842749a71bfbe21afd15edd62498446ee4d8243dd67bf6837a79616b","target":"record","created_at":"2026-05-18T00:03:44Z","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":"7ea83a9c95c097dc1bf4d5c6f2fb67c234f5cacc2817543d76ff3aaa14e3cc04","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-10-09T03:37:08Z","title_canon_sha256":"80055246e3acb0b3c223b9eea6b03117215f753c69bcfc29d13287b394f42a96"},"schema_version":"1.0","source":{"id":"1810.03798","kind":"arxiv","version":1}},"canonical_sha256":"ffb75b4ec66e9f602efc1e16403f953339db20dd7a64c5655c28f348f4da4cec","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ffb75b4ec66e9f602efc1e16403f953339db20dd7a64c5655c28f348f4da4cec","first_computed_at":"2026-05-18T00:03:44.492143Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:03:44.492143Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"tU1ieQ6Iy1zmkzfnS05tOK+n2wr1dzQ+MGeZ7wQZTNqffhB8nAv5Nd6R7KteYG4LMuStUzxKJGsl2av3ssYHAQ==","signature_status":"signed_v1","signed_at":"2026-05-18T00:03:44.492581Z","signed_message":"canonical_sha256_bytes"},"source_id":"1810.03798","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0e9ac183842749a71bfbe21afd15edd62498446ee4d8243dd67bf6837a79616b","sha256:8983e83bba4d784dba010f41b32a9cf5e78e5dc6e7030b9c92152ce38b30625b"],"state_sha256":"5ea94182efebc76ff7128915f5f6b262134782a800f551e99556dcd91d6eb8be"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"83RVRVcYMH2NjnopAcdbEviJFkmPw3ZnCrPNCtj2U5v5G3RR8FCUNnt1vED05muQFnid3U7pPGdQ0W1LA86uCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-28T22:09:56.668095Z","bundle_sha256":"170e37ef354531499c0aa061fb06531713f072aea525aded6d2733dda52ab2bc"}}