{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:MHLU6MWIT4D6GZSJV7EH7QZNIL","short_pith_number":"pith:MHLU6MWI","schema_version":"1.0","canonical_sha256":"61d74f32c89f07e36649afc87fc32d42d8762f9041fab85e0bbd18a0d3265d94","source":{"kind":"arxiv","id":"2606.28573","version":1},"attestation_state":"computed","paper":{"title":"Replica Symmetry Breaking and Algorithmic Thresholds in Empirical Risk Minimization under Multi-Index Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.ST","stat.TH"],"primary_cat":"cs.LG","authors_text":"Andrea Montanari, Kangjie Zhou","submitted_at":"2026-06-26T19:57:08Z","abstract_excerpt":"Modern machine learning models are trained by optimizing high-dimensional non-convex empirical risk functions. Such cost functions can have a multitude of local optima and yet, gradient-based optimization appears to converge to near-global optima.\n  Within a simple supervised learning setting, we develop a precise picture of which parts of the empirical risk landscape are accessible by polynomial-time algorithms. We are given i.i.d. pairs $\\{(\\boldsymbol{x}_i,y_i):\\; 1 \\le i\\le n\\}$ with $\\boldsymbol{x}_i\\in \\mathbb{R}^d$ standard Gaussian feature vectors, and $y_i\\in\\mathbb{R}$ response varia"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.28573","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-06-26T19:57:08Z","cross_cats_sorted":["math.ST","stat.TH"],"title_canon_sha256":"7acbe0346de5ba1d17f676d72f7ab2851975daef70b3b1a7c2269e59aadd9d62","abstract_canon_sha256":"740dadf286db3ed577503dca3f21ee1f609620919e6c9a3e2567e77904170697"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T00:15:19.553569Z","signature_b64":"6LkA5aV3Pfs0je85tjB1lUhjW2e7HWDr0Wqdims8Af7teaqY4nd6sIAjPBd6bIWUcpoqZaxZzpwDVb5V6F/LBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"61d74f32c89f07e36649afc87fc32d42d8762f9041fab85e0bbd18a0d3265d94","last_reissued_at":"2026-06-30T00:15:19.553173Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T00:15:19.553173Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Replica Symmetry Breaking and Algorithmic Thresholds in Empirical Risk Minimization under Multi-Index Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.ST","stat.TH"],"primary_cat":"cs.LG","authors_text":"Andrea Montanari, Kangjie Zhou","submitted_at":"2026-06-26T19:57:08Z","abstract_excerpt":"Modern machine learning models are trained by optimizing high-dimensional non-convex empirical risk functions. Such cost functions can have a multitude of local optima and yet, gradient-based optimization appears to converge to near-global optima.\n  Within a simple supervised learning setting, we develop a precise picture of which parts of the empirical risk landscape are accessible by polynomial-time algorithms. We are given i.i.d. pairs $\\{(\\boldsymbol{x}_i,y_i):\\; 1 \\le i\\le n\\}$ with $\\boldsymbol{x}_i\\in \\mathbb{R}^d$ standard Gaussian feature vectors, and $y_i\\in\\mathbb{R}$ response varia"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.28573","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2606.28573/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.28573","created_at":"2026-06-30T00:15:19.553230+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.28573v1","created_at":"2026-06-30T00:15:19.553230+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.28573","created_at":"2026-06-30T00:15:19.553230+00:00"},{"alias_kind":"pith_short_12","alias_value":"MHLU6MWIT4D6","created_at":"2026-06-30T00:15:19.553230+00:00"},{"alias_kind":"pith_short_16","alias_value":"MHLU6MWIT4D6GZSJ","created_at":"2026-06-30T00:15:19.553230+00:00"},{"alias_kind":"pith_short_8","alias_value":"MHLU6MWI","created_at":"2026-06-30T00:15:19.553230+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MHLU6MWIT4D6GZSJV7EH7QZNIL","json":"https://pith.science/pith/MHLU6MWIT4D6GZSJV7EH7QZNIL.json","graph_json":"https://pith.science/api/pith-number/MHLU6MWIT4D6GZSJV7EH7QZNIL/graph.json","events_json":"https://pith.science/api/pith-number/MHLU6MWIT4D6GZSJV7EH7QZNIL/events.json","paper":"https://pith.science/paper/MHLU6MWI"},"agent_actions":{"view_html":"https://pith.science/pith/MHLU6MWIT4D6GZSJV7EH7QZNIL","download_json":"https://pith.science/pith/MHLU6MWIT4D6GZSJV7EH7QZNIL.json","view_paper":"https://pith.science/paper/MHLU6MWI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.28573&json=true","fetch_graph":"https://pith.science/api/pith-number/MHLU6MWIT4D6GZSJV7EH7QZNIL/graph.json","fetch_events":"https://pith.science/api/pith-number/MHLU6MWIT4D6GZSJV7EH7QZNIL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MHLU6MWIT4D6GZSJV7EH7QZNIL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MHLU6MWIT4D6GZSJV7EH7QZNIL/action/storage_attestation","attest_author":"https://pith.science/pith/MHLU6MWIT4D6GZSJV7EH7QZNIL/action/author_attestation","sign_citation":"https://pith.science/pith/MHLU6MWIT4D6GZSJV7EH7QZNIL/action/citation_signature","submit_replication":"https://pith.science/pith/MHLU6MWIT4D6GZSJV7EH7QZNIL/action/replication_record"}},"created_at":"2026-06-30T00:15:19.553230+00:00","updated_at":"2026-06-30T00:15:19.553230+00:00"}