{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:FWMY3CMYDEXHCHULCLDD7NIYLM","short_pith_number":"pith:FWMY3CMY","canonical_record":{"source":{"id":"1710.10174","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-10-27T14:42:09Z","cross_cats_sorted":[],"title_canon_sha256":"abc704d43ca2d27e446733f68d509fec3bc0b4d36caa30006165c45468e39c97","abstract_canon_sha256":"eba065ba70ab1004ae42f0418525503dee28c7101e2c4d6391e18b2933f13949"},"schema_version":"1.0"},"canonical_sha256":"2d998d8998192e711e8b12c63fb5185b1ce67bdd89001c63a219368b843570f2","source":{"kind":"arxiv","id":"1710.10174","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1710.10174","created_at":"2026-05-18T00:31:54Z"},{"alias_kind":"arxiv_version","alias_value":"1710.10174v1","created_at":"2026-05-18T00:31:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1710.10174","created_at":"2026-05-18T00:31:54Z"},{"alias_kind":"pith_short_12","alias_value":"FWMY3CMYDEXH","created_at":"2026-05-18T12:31:15Z"},{"alias_kind":"pith_short_16","alias_value":"FWMY3CMYDEXHCHUL","created_at":"2026-05-18T12:31:15Z"},{"alias_kind":"pith_short_8","alias_value":"FWMY3CMY","created_at":"2026-05-18T12:31:15Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:FWMY3CMYDEXHCHULCLDD7NIYLM","target":"record","payload":{"canonical_record":{"source":{"id":"1710.10174","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-10-27T14:42:09Z","cross_cats_sorted":[],"title_canon_sha256":"abc704d43ca2d27e446733f68d509fec3bc0b4d36caa30006165c45468e39c97","abstract_canon_sha256":"eba065ba70ab1004ae42f0418525503dee28c7101e2c4d6391e18b2933f13949"},"schema_version":"1.0"},"canonical_sha256":"2d998d8998192e711e8b12c63fb5185b1ce67bdd89001c63a219368b843570f2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:31:54.601387Z","signature_b64":"Lz24h7tSP5F8e6z0Q0CqF22Cz1fx+b3Arw43AZSz5QPUnAMkYqaZhakNxNUGXdqZa6rsigRGZwSLU7T15UDQCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2d998d8998192e711e8b12c63fb5185b1ce67bdd89001c63a219368b843570f2","last_reissued_at":"2026-05-18T00:31:54.600757Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:31:54.600757Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1710.10174","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":"2026-05-18T00:31:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DQ6kPMU3nmQcvNlzHkDdyAaVOnHuy5xopuoWIEPmtRLjMiPC87vRXdjyrwFHkBUsDPcRlONgYlH8P148ch++Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T09:48:34.342409Z"},"content_sha256":"699ceef654eb95273770176b8fac2ae9c6deb04d90e78fab5ab6c52ab1977e01","schema_version":"1.0","event_id":"sha256:699ceef654eb95273770176b8fac2ae9c6deb04d90e78fab5ab6c52ab1977e01"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:FWMY3CMYDEXHCHULCLDD7NIYLM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SGD Learns Over-parameterized Networks that Provably Generalize on Linearly Separable Data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alon Brutzkus, Amir Globerson, Eran Malach, Shai Shalev-Shwartz","submitted_at":"2017-10-27T14:42:09Z","abstract_excerpt":"Neural networks exhibit good generalization behavior in the over-parameterized regime, where the number of network parameters exceeds the number of observations. Nonetheless, current generalization bounds for neural networks fail to explain this phenomenon. In an attempt to bridge this gap, we study the problem of learning a two-layer over-parameterized neural network, when the data is generated by a linearly separable function. In the case where the network has Leaky ReLU activations, we provide both optimization and generalization guarantees for over-parameterized networks. Specifically, we "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1710.10174","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":"2026-05-18T00:31:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ya7yPcHbGRInu5r0U5z+xKipSLlX6WOC2E9ZqlWBAaPxkRVBIxa63uub7Pz5ib9i33LltcEHG8tTwzAwpdk0Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T09:48:34.342993Z"},"content_sha256":"cd5bd9b88cf9f48051946f696660845d004d6d46cd2fe2fb7e6883464674d025","schema_version":"1.0","event_id":"sha256:cd5bd9b88cf9f48051946f696660845d004d6d46cd2fe2fb7e6883464674d025"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FWMY3CMYDEXHCHULCLDD7NIYLM/bundle.json","state_url":"https://pith.science/pith/FWMY3CMYDEXHCHULCLDD7NIYLM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FWMY3CMYDEXHCHULCLDD7NIYLM/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":"2026-05-25T09:48:34Z","links":{"resolver":"https://pith.science/pith/FWMY3CMYDEXHCHULCLDD7NIYLM","bundle":"https://pith.science/pith/FWMY3CMYDEXHCHULCLDD7NIYLM/bundle.json","state":"https://pith.science/pith/FWMY3CMYDEXHCHULCLDD7NIYLM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FWMY3CMYDEXHCHULCLDD7NIYLM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:FWMY3CMYDEXHCHULCLDD7NIYLM","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":"eba065ba70ab1004ae42f0418525503dee28c7101e2c4d6391e18b2933f13949","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-10-27T14:42:09Z","title_canon_sha256":"abc704d43ca2d27e446733f68d509fec3bc0b4d36caa30006165c45468e39c97"},"schema_version":"1.0","source":{"id":"1710.10174","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1710.10174","created_at":"2026-05-18T00:31:54Z"},{"alias_kind":"arxiv_version","alias_value":"1710.10174v1","created_at":"2026-05-18T00:31:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1710.10174","created_at":"2026-05-18T00:31:54Z"},{"alias_kind":"pith_short_12","alias_value":"FWMY3CMYDEXH","created_at":"2026-05-18T12:31:15Z"},{"alias_kind":"pith_short_16","alias_value":"FWMY3CMYDEXHCHUL","created_at":"2026-05-18T12:31:15Z"},{"alias_kind":"pith_short_8","alias_value":"FWMY3CMY","created_at":"2026-05-18T12:31:15Z"}],"graph_snapshots":[{"event_id":"sha256:cd5bd9b88cf9f48051946f696660845d004d6d46cd2fe2fb7e6883464674d025","target":"graph","created_at":"2026-05-18T00:31:54Z","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":"Neural networks exhibit good generalization behavior in the over-parameterized regime, where the number of network parameters exceeds the number of observations. Nonetheless, current generalization bounds for neural networks fail to explain this phenomenon. In an attempt to bridge this gap, we study the problem of learning a two-layer over-parameterized neural network, when the data is generated by a linearly separable function. In the case where the network has Leaky ReLU activations, we provide both optimization and generalization guarantees for over-parameterized networks. Specifically, we ","authors_text":"Alon Brutzkus, Amir Globerson, Eran Malach, Shai Shalev-Shwartz","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-10-27T14:42:09Z","title":"SGD Learns Over-parameterized Networks that Provably Generalize on Linearly Separable Data"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1710.10174","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":"sha256:699ceef654eb95273770176b8fac2ae9c6deb04d90e78fab5ab6c52ab1977e01","target":"record","created_at":"2026-05-18T00:31:54Z","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":"eba065ba70ab1004ae42f0418525503dee28c7101e2c4d6391e18b2933f13949","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-10-27T14:42:09Z","title_canon_sha256":"abc704d43ca2d27e446733f68d509fec3bc0b4d36caa30006165c45468e39c97"},"schema_version":"1.0","source":{"id":"1710.10174","kind":"arxiv","version":1}},"canonical_sha256":"2d998d8998192e711e8b12c63fb5185b1ce67bdd89001c63a219368b843570f2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2d998d8998192e711e8b12c63fb5185b1ce67bdd89001c63a219368b843570f2","first_computed_at":"2026-05-18T00:31:54.600757Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:31:54.600757Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Lz24h7tSP5F8e6z0Q0CqF22Cz1fx+b3Arw43AZSz5QPUnAMkYqaZhakNxNUGXdqZa6rsigRGZwSLU7T15UDQCQ==","signature_status":"signed_v1","signed_at":"2026-05-18T00:31:54.601387Z","signed_message":"canonical_sha256_bytes"},"source_id":"1710.10174","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:699ceef654eb95273770176b8fac2ae9c6deb04d90e78fab5ab6c52ab1977e01","sha256:cd5bd9b88cf9f48051946f696660845d004d6d46cd2fe2fb7e6883464674d025"],"state_sha256":"755988fded8d80f18d57b51f60b12c9db922c203a6b67dc5680c6ce3afbb1a9b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BRD26RIjyhIy4oBk0RyKRjbeyk0Iz9zO8DOVlQBQe1tUbFZtPi1/NJxCKVQ52brHlZw38jOEj5OWJglTS6NNDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-25T09:48:34.345811Z","bundle_sha256":"d7e4c957aa7dee17dcf5a2fa4805e363b09ebacbfd651fd335209fcac5a974b7"}}