{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:3JNLGQBZNTTPZZARRSBLEXXGGQ","short_pith_number":"pith:3JNLGQBZ","canonical_record":{"source":{"id":"2305.18502","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2023-05-29T14:40:56Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"c3422caa45f25052b14d26b31d1f854e13d3ab785aad226add5befaf4caa1dfa","abstract_canon_sha256":"c220bc74e99062bb8896016fb0591632addc43a3998af6614741cbf3131b370b"},"schema_version":"1.0"},"canonical_sha256":"da5ab340396ce6fce4118c82b25ee6342eb3eae9837668258ced95ef19687268","source":{"kind":"arxiv","id":"2305.18502","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.18502","created_at":"2026-07-05T07:50:50Z"},{"alias_kind":"arxiv_version","alias_value":"2305.18502v2","created_at":"2026-07-05T07:50:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.18502","created_at":"2026-07-05T07:50:50Z"},{"alias_kind":"pith_short_12","alias_value":"3JNLGQBZNTTP","created_at":"2026-07-05T07:50:50Z"},{"alias_kind":"pith_short_16","alias_value":"3JNLGQBZNTTPZZAR","created_at":"2026-07-05T07:50:50Z"},{"alias_kind":"pith_short_8","alias_value":"3JNLGQBZ","created_at":"2026-07-05T07:50:50Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:3JNLGQBZNTTPZZARRSBLEXXGGQ","target":"record","payload":{"canonical_record":{"source":{"id":"2305.18502","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2023-05-29T14:40:56Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"c3422caa45f25052b14d26b31d1f854e13d3ab785aad226add5befaf4caa1dfa","abstract_canon_sha256":"c220bc74e99062bb8896016fb0591632addc43a3998af6614741cbf3131b370b"},"schema_version":"1.0"},"canonical_sha256":"da5ab340396ce6fce4118c82b25ee6342eb3eae9837668258ced95ef19687268","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:50:50.085229Z","signature_b64":"gihol9NlVEVMT9mhN8U+JuvxTYNSziHjbHuwT1iE/EcA+t4GEq1Kc0MKbV718MAFEJYTgdhFPWZC25YeYu/0AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"da5ab340396ce6fce4118c82b25ee6342eb3eae9837668258ced95ef19687268","last_reissued_at":"2026-07-05T07:50:50.084721Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:50:50.084721Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2305.18502","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:50:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pzjofPpSUIZNzS7nRS+ZXzs7pr1e5xCUckyPsFJ8IrtXx06rzuIbSYlWFAxJlDdrlwhacTUANpKuxULxoae7Aw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T08:23:59.667388Z"},"content_sha256":"432f9a999a3e02a644c2f76bb5ef09dd12a55db00baba3c4c44f64e405057e70","schema_version":"1.0","event_id":"sha256:432f9a999a3e02a644c2f76bb5ef09dd12a55db00baba3c4c44f64e405057e70"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:3JNLGQBZNTTPZZARRSBLEXXGGQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Escaping mediocrity: how two-layer networks learn hard generalized linear models with SGD","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Bruno Loureiro, Florent Krzakala, Luca Arnaboldi, Ludovic Stephan","submitted_at":"2023-05-29T14:40:56Z","abstract_excerpt":"This study explores the sample complexity for two-layer neural networks to learn a generalized linear target function under Stochastic Gradient Descent (SGD), focusing on the challenging regime where many flat directions are present at initialization. It is well-established that in this scenario $n=O(d \\log d)$ samples are typically needed. However, we provide precise results concerning the pre-factors in high-dimensional contexts and for varying widths. Notably, our findings suggest that overparameterization can only enhance convergence by a constant factor within this problem class. These in"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.18502","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2305.18502/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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":"2026-07-05T07:50:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6AvJMy9pZGYQiMcY8qsKLE+kDWhL3+eLJTxbLbUK9AcJFd7o/+PYPgy/jlSuiKGXOyLm9RqwqBX6Amx6hl3SAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T08:23:59.667880Z"},"content_sha256":"e586362d8a802217a1ed7b2f6710cdca846577b2f7c8c636dd415891c4c814ab","schema_version":"1.0","event_id":"sha256:e586362d8a802217a1ed7b2f6710cdca846577b2f7c8c636dd415891c4c814ab"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/3JNLGQBZNTTPZZARRSBLEXXGGQ/bundle.json","state_url":"https://pith.science/pith/3JNLGQBZNTTPZZARRSBLEXXGGQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/3JNLGQBZNTTPZZARRSBLEXXGGQ/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":"2026-08-11T08:23:59Z","links":{"resolver":"https://pith.science/pith/3JNLGQBZNTTPZZARRSBLEXXGGQ","bundle":"https://pith.science/pith/3JNLGQBZNTTPZZARRSBLEXXGGQ/bundle.json","state":"https://pith.science/pith/3JNLGQBZNTTPZZARRSBLEXXGGQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/3JNLGQBZNTTPZZARRSBLEXXGGQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:3JNLGQBZNTTPZZARRSBLEXXGGQ","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":"c220bc74e99062bb8896016fb0591632addc43a3998af6614741cbf3131b370b","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2023-05-29T14:40:56Z","title_canon_sha256":"c3422caa45f25052b14d26b31d1f854e13d3ab785aad226add5befaf4caa1dfa"},"schema_version":"1.0","source":{"id":"2305.18502","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.18502","created_at":"2026-07-05T07:50:50Z"},{"alias_kind":"arxiv_version","alias_value":"2305.18502v2","created_at":"2026-07-05T07:50:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.18502","created_at":"2026-07-05T07:50:50Z"},{"alias_kind":"pith_short_12","alias_value":"3JNLGQBZNTTP","created_at":"2026-07-05T07:50:50Z"},{"alias_kind":"pith_short_16","alias_value":"3JNLGQBZNTTPZZAR","created_at":"2026-07-05T07:50:50Z"},{"alias_kind":"pith_short_8","alias_value":"3JNLGQBZ","created_at":"2026-07-05T07:50:50Z"}],"graph_snapshots":[{"event_id":"sha256:e586362d8a802217a1ed7b2f6710cdca846577b2f7c8c636dd415891c4c814ab","target":"graph","created_at":"2026-07-05T07:50:50Z","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"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2305.18502/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This study explores the sample complexity for two-layer neural networks to learn a generalized linear target function under Stochastic Gradient Descent (SGD), focusing on the challenging regime where many flat directions are present at initialization. It is well-established that in this scenario $n=O(d \\log d)$ samples are typically needed. However, we provide precise results concerning the pre-factors in high-dimensional contexts and for varying widths. Notably, our findings suggest that overparameterization can only enhance convergence by a constant factor within this problem class. These in","authors_text":"Bruno Loureiro, Florent Krzakala, Luca Arnaboldi, Ludovic Stephan","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2023-05-29T14:40:56Z","title":"Escaping mediocrity: how two-layer networks learn hard generalized linear models with SGD"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.18502","kind":"arxiv","version":2},"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":"sha256:432f9a999a3e02a644c2f76bb5ef09dd12a55db00baba3c4c44f64e405057e70","target":"record","created_at":"2026-07-05T07:50:50Z","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":"c220bc74e99062bb8896016fb0591632addc43a3998af6614741cbf3131b370b","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2023-05-29T14:40:56Z","title_canon_sha256":"c3422caa45f25052b14d26b31d1f854e13d3ab785aad226add5befaf4caa1dfa"},"schema_version":"1.0","source":{"id":"2305.18502","kind":"arxiv","version":2}},"canonical_sha256":"da5ab340396ce6fce4118c82b25ee6342eb3eae9837668258ced95ef19687268","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"da5ab340396ce6fce4118c82b25ee6342eb3eae9837668258ced95ef19687268","first_computed_at":"2026-07-05T07:50:50.084721Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:50:50.084721Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"gihol9NlVEVMT9mhN8U+JuvxTYNSziHjbHuwT1iE/EcA+t4GEq1Kc0MKbV718MAFEJYTgdhFPWZC25YeYu/0AA==","signature_status":"signed_v1","signed_at":"2026-07-05T07:50:50.085229Z","signed_message":"canonical_sha256_bytes"},"source_id":"2305.18502","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:432f9a999a3e02a644c2f76bb5ef09dd12a55db00baba3c4c44f64e405057e70","sha256:e586362d8a802217a1ed7b2f6710cdca846577b2f7c8c636dd415891c4c814ab"],"state_sha256":"874a9a233b099346ec7aa04debba8ea75397a910d8ecabdcd095bb2a85578c93"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"B8i5HCTsDGbNipaQcVbRGP88+x4ycw8dtNgnoR1R1RvP0rqtcX1T/SLK+ZR5re/b2P5hPmZahzDCL+ofMLLFBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T08:23:59.672650Z","bundle_sha256":"b5ca410d41515a6d50ad970c41b0a195841d941a0e5ec0b937353edda96bef6b"}}