{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:KOLTI2OFYHR2ZYQBUHC3EJSLLX","short_pith_number":"pith:KOLTI2OF","schema_version":"1.0","canonical_sha256":"53973469c5c1e3ace201a1c5b2264b5df8b4489d7626ef8c0d9afc34ea7f4852","source":{"kind":"arxiv","id":"2410.03477","version":1},"attestation_state":"computed","paper":{"title":"On the Hardness of Learning One Hidden Layer Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CC","math.ST","stat.ML","stat.TH"],"primary_cat":"cs.LG","authors_text":"Ilias Zadik, Manolis Zampetakis, Shuchen Li","submitted_at":"2024-10-04T14:48:13Z","abstract_excerpt":"In this work, we consider the problem of learning one hidden layer ReLU neural networks with inputs from $\\mathbb{R}^d$. We show that this learning problem is hard under standard cryptographic assumptions even when: (1) the size of the neural network is polynomial in $d$, (2) its input distribution is a standard Gaussian, and (3) the noise is Gaussian and polynomially small in $d$. Our hardness result is based on the hardness of the Continuous Learning with Errors (CLWE) problem, and in particular, is based on the largely believed worst-case hardness of approximately solving the shortest vecto"},"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":"2410.03477","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-10-04T14:48:13Z","cross_cats_sorted":["cs.CC","math.ST","stat.ML","stat.TH"],"title_canon_sha256":"209664e20b83e78fd5e9397f4f91a19591c40e52bddd52f381932eb5cabcb251","abstract_canon_sha256":"9c5e1bf0a43a57c37d5ba17972f950303dddc8596e3b97f6ded4515ccadfc86b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:16:01.717926Z","signature_b64":"LLsXmASnsCgt2yeW9Be7eJz01QfbsaFKH2Z2COgUU/HL8kh0pfv7N/55AlubB6ZYBxcvLgC6zrGMQo/iBc6EBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"53973469c5c1e3ace201a1c5b2264b5df8b4489d7626ef8c0d9afc34ea7f4852","last_reissued_at":"2026-07-05T09:16:01.717459Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:16:01.717459Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On the Hardness of Learning One Hidden Layer Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CC","math.ST","stat.ML","stat.TH"],"primary_cat":"cs.LG","authors_text":"Ilias Zadik, Manolis Zampetakis, Shuchen Li","submitted_at":"2024-10-04T14:48:13Z","abstract_excerpt":"In this work, we consider the problem of learning one hidden layer ReLU neural networks with inputs from $\\mathbb{R}^d$. We show that this learning problem is hard under standard cryptographic assumptions even when: (1) the size of the neural network is polynomial in $d$, (2) its input distribution is a standard Gaussian, and (3) the noise is Gaussian and polynomially small in $d$. Our hardness result is based on the hardness of the Continuous Learning with Errors (CLWE) problem, and in particular, is based on the largely believed worst-case hardness of approximately solving the shortest vecto"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.03477","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/2410.03477/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":"2410.03477","created_at":"2026-07-05T09:16:01.717513+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.03477v1","created_at":"2026-07-05T09:16:01.717513+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.03477","created_at":"2026-07-05T09:16:01.717513+00:00"},{"alias_kind":"pith_short_12","alias_value":"KOLTI2OFYHR2","created_at":"2026-07-05T09:16:01.717513+00:00"},{"alias_kind":"pith_short_16","alias_value":"KOLTI2OFYHR2ZYQB","created_at":"2026-07-05T09:16:01.717513+00:00"},{"alias_kind":"pith_short_8","alias_value":"KOLTI2OF","created_at":"2026-07-05T09:16:01.717513+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.15669","citing_title":"Implicit High-Order Moment Tensor Estimation and Learning Latent Variable Models","ref_index":35,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KOLTI2OFYHR2ZYQBUHC3EJSLLX","json":"https://pith.science/pith/KOLTI2OFYHR2ZYQBUHC3EJSLLX.json","graph_json":"https://pith.science/api/pith-number/KOLTI2OFYHR2ZYQBUHC3EJSLLX/graph.json","events_json":"https://pith.science/api/pith-number/KOLTI2OFYHR2ZYQBUHC3EJSLLX/events.json","paper":"https://pith.science/paper/KOLTI2OF"},"agent_actions":{"view_html":"https://pith.science/pith/KOLTI2OFYHR2ZYQBUHC3EJSLLX","download_json":"https://pith.science/pith/KOLTI2OFYHR2ZYQBUHC3EJSLLX.json","view_paper":"https://pith.science/paper/KOLTI2OF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.03477&json=true","fetch_graph":"https://pith.science/api/pith-number/KOLTI2OFYHR2ZYQBUHC3EJSLLX/graph.json","fetch_events":"https://pith.science/api/pith-number/KOLTI2OFYHR2ZYQBUHC3EJSLLX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KOLTI2OFYHR2ZYQBUHC3EJSLLX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KOLTI2OFYHR2ZYQBUHC3EJSLLX/action/storage_attestation","attest_author":"https://pith.science/pith/KOLTI2OFYHR2ZYQBUHC3EJSLLX/action/author_attestation","sign_citation":"https://pith.science/pith/KOLTI2OFYHR2ZYQBUHC3EJSLLX/action/citation_signature","submit_replication":"https://pith.science/pith/KOLTI2OFYHR2ZYQBUHC3EJSLLX/action/replication_record"}},"created_at":"2026-07-05T09:16:01.717513+00:00","updated_at":"2026-07-05T09:16:01.717513+00:00"}