{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:MVHS7BIG6LW4KL6JDVQG67MYG3","short_pith_number":"pith:MVHS7BIG","canonical_record":{"source":{"id":"1712.09203","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-12-26T08:04:43Z","cross_cats_sorted":["cs.DS","math.OC","stat.ML"],"title_canon_sha256":"d14addae24dc3a7a554ea2e6add3264a7bdec192ee8c087c47e5f0c13c357886","abstract_canon_sha256":"d2c7e950feb19efe61230fd4a01b8f8773ed1751f6a05292373da92e559fc9a4"},"schema_version":"1.0"},"canonical_sha256":"654f2f8506f2edc52fc91d606f7d9836e67ccb29c162b84a5de77ee124dc0a7b","source":{"kind":"arxiv","id":"1712.09203","version":5},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1712.09203","created_at":"2026-05-17T23:54:03Z"},{"alias_kind":"arxiv_version","alias_value":"1712.09203v5","created_at":"2026-05-17T23:54:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1712.09203","created_at":"2026-05-17T23:54:03Z"},{"alias_kind":"pith_short_12","alias_value":"MVHS7BIG6LW4","created_at":"2026-05-18T12:31:31Z"},{"alias_kind":"pith_short_16","alias_value":"MVHS7BIG6LW4KL6J","created_at":"2026-05-18T12:31:31Z"},{"alias_kind":"pith_short_8","alias_value":"MVHS7BIG","created_at":"2026-05-18T12:31:31Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:MVHS7BIG6LW4KL6JDVQG67MYG3","target":"record","payload":{"canonical_record":{"source":{"id":"1712.09203","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-12-26T08:04:43Z","cross_cats_sorted":["cs.DS","math.OC","stat.ML"],"title_canon_sha256":"d14addae24dc3a7a554ea2e6add3264a7bdec192ee8c087c47e5f0c13c357886","abstract_canon_sha256":"d2c7e950feb19efe61230fd4a01b8f8773ed1751f6a05292373da92e559fc9a4"},"schema_version":"1.0"},"canonical_sha256":"654f2f8506f2edc52fc91d606f7d9836e67ccb29c162b84a5de77ee124dc0a7b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:54:03.141861Z","signature_b64":"1AnLbLdMFibE5DTQy0x1L2QJ970WQeAF1NNvsmIRy90kVErivyOfFUdf3gmBlsfAvNBYlrXZVOJ3twbA3Xb1AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"654f2f8506f2edc52fc91d606f7d9836e67ccb29c162b84a5de77ee124dc0a7b","last_reissued_at":"2026-05-17T23:54:03.141245Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:54:03.141245Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1712.09203","source_version":5,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:54:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"G7gdBHoSIewh8JjllCg2vYt64s9TGLGxp2/xwOAHs77WJ36vcsuQzdCOP8/0+I7ddyVKlHOTcGNJX5Rjdh6TDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-03T16:13:45.404449Z"},"content_sha256":"fa1f9ea56ac6a4f605438c887a486fa34ee2f55158cb293f9f7d7ed304762dc2","schema_version":"1.0","event_id":"sha256:fa1f9ea56ac6a4f605438c887a486fa34ee2f55158cb293f9f7d7ed304762dc2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:MVHS7BIG6LW4KL6JDVQG67MYG3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Algorithmic Regularization in Over-parameterized Matrix Sensing and Neural Networks with Quadratic Activations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DS","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Hongyang Zhang, Tengyu Ma, Yuanzhi Li","submitted_at":"2017-12-26T08:04:43Z","abstract_excerpt":"We show that the gradient descent algorithm provides an implicit regularization effect in the learning of over-parameterized matrix factorization models and one-hidden-layer neural networks with quadratic activations. Concretely, we show that given $\\tilde{O}(dr^{2})$ random linear measurements of a rank $r$ positive semidefinite matrix $X^{\\star}$, we can recover $X^{\\star}$ by parameterizing it by $UU^\\top$ with $U\\in \\mathbb R^{d\\times d}$ and minimizing the squared loss, even if $r \\ll d$. We prove that starting from a small initialization, gradient descent recovers $X^{\\star}$ in $\\tilde{"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1712.09203","kind":"arxiv","version":5},"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":"2026-05-17T23:54:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Xs1sIzdwIRR5XV9RqPqhMbsO5rEsPaf/j4XFIIAOVkB+GYLOkuhfTJaTpEJoDkcYt8tgWOIjFZmDAd3YuBfUBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-03T16:13:45.404800Z"},"content_sha256":"b303a8c7e96cd97638df1cb1482e97cf0df4e8cf5096d2b8a745900fba03bc0e","schema_version":"1.0","event_id":"sha256:b303a8c7e96cd97638df1cb1482e97cf0df4e8cf5096d2b8a745900fba03bc0e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/MVHS7BIG6LW4KL6JDVQG67MYG3/bundle.json","state_url":"https://pith.science/pith/MVHS7BIG6LW4KL6JDVQG67MYG3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/MVHS7BIG6LW4KL6JDVQG67MYG3/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":"2026-06-03T16:13:45Z","links":{"resolver":"https://pith.science/pith/MVHS7BIG6LW4KL6JDVQG67MYG3","bundle":"https://pith.science/pith/MVHS7BIG6LW4KL6JDVQG67MYG3/bundle.json","state":"https://pith.science/pith/MVHS7BIG6LW4KL6JDVQG67MYG3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/MVHS7BIG6LW4KL6JDVQG67MYG3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:MVHS7BIG6LW4KL6JDVQG67MYG3","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":"d2c7e950feb19efe61230fd4a01b8f8773ed1751f6a05292373da92e559fc9a4","cross_cats_sorted":["cs.DS","math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-12-26T08:04:43Z","title_canon_sha256":"d14addae24dc3a7a554ea2e6add3264a7bdec192ee8c087c47e5f0c13c357886"},"schema_version":"1.0","source":{"id":"1712.09203","kind":"arxiv","version":5}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1712.09203","created_at":"2026-05-17T23:54:03Z"},{"alias_kind":"arxiv_version","alias_value":"1712.09203v5","created_at":"2026-05-17T23:54:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1712.09203","created_at":"2026-05-17T23:54:03Z"},{"alias_kind":"pith_short_12","alias_value":"MVHS7BIG6LW4","created_at":"2026-05-18T12:31:31Z"},{"alias_kind":"pith_short_16","alias_value":"MVHS7BIG6LW4KL6J","created_at":"2026-05-18T12:31:31Z"},{"alias_kind":"pith_short_8","alias_value":"MVHS7BIG","created_at":"2026-05-18T12:31:31Z"}],"graph_snapshots":[{"event_id":"sha256:b303a8c7e96cd97638df1cb1482e97cf0df4e8cf5096d2b8a745900fba03bc0e","target":"graph","created_at":"2026-05-17T23:54:03Z","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":"We show that the gradient descent algorithm provides an implicit regularization effect in the learning of over-parameterized matrix factorization models and one-hidden-layer neural networks with quadratic activations. Concretely, we show that given $\\tilde{O}(dr^{2})$ random linear measurements of a rank $r$ positive semidefinite matrix $X^{\\star}$, we can recover $X^{\\star}$ by parameterizing it by $UU^\\top$ with $U\\in \\mathbb R^{d\\times d}$ and minimizing the squared loss, even if $r \\ll d$. We prove that starting from a small initialization, gradient descent recovers $X^{\\star}$ in $\\tilde{","authors_text":"Hongyang Zhang, Tengyu Ma, Yuanzhi Li","cross_cats":["cs.DS","math.OC","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-12-26T08:04:43Z","title":"Algorithmic Regularization in Over-parameterized Matrix Sensing and Neural Networks with Quadratic Activations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1712.09203","kind":"arxiv","version":5},"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:fa1f9ea56ac6a4f605438c887a486fa34ee2f55158cb293f9f7d7ed304762dc2","target":"record","created_at":"2026-05-17T23:54:03Z","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":"d2c7e950feb19efe61230fd4a01b8f8773ed1751f6a05292373da92e559fc9a4","cross_cats_sorted":["cs.DS","math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-12-26T08:04:43Z","title_canon_sha256":"d14addae24dc3a7a554ea2e6add3264a7bdec192ee8c087c47e5f0c13c357886"},"schema_version":"1.0","source":{"id":"1712.09203","kind":"arxiv","version":5}},"canonical_sha256":"654f2f8506f2edc52fc91d606f7d9836e67ccb29c162b84a5de77ee124dc0a7b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"654f2f8506f2edc52fc91d606f7d9836e67ccb29c162b84a5de77ee124dc0a7b","first_computed_at":"2026-05-17T23:54:03.141245Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:54:03.141245Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1AnLbLdMFibE5DTQy0x1L2QJ970WQeAF1NNvsmIRy90kVErivyOfFUdf3gmBlsfAvNBYlrXZVOJ3twbA3Xb1AA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:54:03.141861Z","signed_message":"canonical_sha256_bytes"},"source_id":"1712.09203","source_kind":"arxiv","source_version":5}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:fa1f9ea56ac6a4f605438c887a486fa34ee2f55158cb293f9f7d7ed304762dc2","sha256:b303a8c7e96cd97638df1cb1482e97cf0df4e8cf5096d2b8a745900fba03bc0e"],"state_sha256":"87e605ce55eea51c4ca7114d204c45d8bb6244338a2ac1b0abe5dda2e27cf985"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SRap6Dl6uVdk0HakGjP66W8Q+4oiypgY+53YmKZxM4pK9Nnhl+7UseGn08JWGH9zFqjwCxZ7TiOAYP99Ehn6Bw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-03T16:13:45.406799Z","bundle_sha256":"4d5f4b065fe97d875d1842ef665f831b4d5ea3a8c5b00752ba626f95b5f806d9"}}