{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:2ILTPYNY3GF4AEIXSM4TKFTQSF","short_pith_number":"pith:2ILTPYNY","schema_version":"1.0","canonical_sha256":"d21737e1b8d98bc011179339351670914f7b98eeae66c2e60ee511549c273f4e","source":{"kind":"arxiv","id":"1802.03713","version":8},"attestation_state":"computed","paper":{"title":"$\\mathcal{G}$-SGD: Optimizing ReLU Neural Networks in its Positively Scale-Invariant Space","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Huishuai Zhang, Qi Meng, Shuxin Zheng, Tie-Yan Liu, Wei Chen, Zhi-Ming Ma","submitted_at":"2018-02-11T08:57:08Z","abstract_excerpt":"It is well known that neural networks with rectified linear units (ReLU) activation functions are positively scale-invariant. Conventional algorithms like stochastic gradient descent optimize the neural networks in the vector space of weights, which is, however, not positively scale-invariant. This mismatch may lead to problems during the optimization process. Then, a natural question is: \\emph{can we construct a new vector space that is positively scale-invariant and sufficient to represent ReLU neural networks so as to better facilitate the optimization process }? In this paper, we provide o"},"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":"1802.03713","kind":"arxiv","version":8},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2018-02-11T08:57:08Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"662069aabdd2022a038230cc9e7156bf6491755db596da1dacc9a44e4dfd0a34","abstract_canon_sha256":"01fa3e710027d99292265ea490b5aa0a8f39a9c14d26064bd256ea2eddfe12da"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:25:15.034448Z","signature_b64":"aUnYv+NFStL1whHwowZKxA19i2lKTV2QmIK2JP00AquQgBKKseLrgygKHaqhVI+nLb6zAbqtPTHBI25kDcxsDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d21737e1b8d98bc011179339351670914f7b98eeae66c2e60ee511549c273f4e","last_reissued_at":"2026-07-05T02:25:15.033975Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:25:15.033975Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"$\\mathcal{G}$-SGD: Optimizing ReLU Neural Networks in its Positively Scale-Invariant Space","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Huishuai Zhang, Qi Meng, Shuxin Zheng, Tie-Yan Liu, Wei Chen, Zhi-Ming Ma","submitted_at":"2018-02-11T08:57:08Z","abstract_excerpt":"It is well known that neural networks with rectified linear units (ReLU) activation functions are positively scale-invariant. Conventional algorithms like stochastic gradient descent optimize the neural networks in the vector space of weights, which is, however, not positively scale-invariant. This mismatch may lead to problems during the optimization process. Then, a natural question is: \\emph{can we construct a new vector space that is positively scale-invariant and sufficient to represent ReLU neural networks so as to better facilitate the optimization process }? In this paper, we provide o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1802.03713","kind":"arxiv","version":8},"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/1802.03713/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":"1802.03713","created_at":"2026-07-05T02:25:15.034034+00:00"},{"alias_kind":"arxiv_version","alias_value":"1802.03713v8","created_at":"2026-07-05T02:25:15.034034+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1802.03713","created_at":"2026-07-05T02:25:15.034034+00:00"},{"alias_kind":"pith_short_12","alias_value":"2ILTPYNY3GF4","created_at":"2026-07-05T02:25:15.034034+00:00"},{"alias_kind":"pith_short_16","alias_value":"2ILTPYNY3GF4AEIX","created_at":"2026-07-05T02:25:15.034034+00:00"},{"alias_kind":"pith_short_8","alias_value":"2ILTPYNY","created_at":"2026-07-05T02:25:15.034034+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/2ILTPYNY3GF4AEIXSM4TKFTQSF","json":"https://pith.science/pith/2ILTPYNY3GF4AEIXSM4TKFTQSF.json","graph_json":"https://pith.science/api/pith-number/2ILTPYNY3GF4AEIXSM4TKFTQSF/graph.json","events_json":"https://pith.science/api/pith-number/2ILTPYNY3GF4AEIXSM4TKFTQSF/events.json","paper":"https://pith.science/paper/2ILTPYNY"},"agent_actions":{"view_html":"https://pith.science/pith/2ILTPYNY3GF4AEIXSM4TKFTQSF","download_json":"https://pith.science/pith/2ILTPYNY3GF4AEIXSM4TKFTQSF.json","view_paper":"https://pith.science/paper/2ILTPYNY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1802.03713&json=true","fetch_graph":"https://pith.science/api/pith-number/2ILTPYNY3GF4AEIXSM4TKFTQSF/graph.json","fetch_events":"https://pith.science/api/pith-number/2ILTPYNY3GF4AEIXSM4TKFTQSF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2ILTPYNY3GF4AEIXSM4TKFTQSF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2ILTPYNY3GF4AEIXSM4TKFTQSF/action/storage_attestation","attest_author":"https://pith.science/pith/2ILTPYNY3GF4AEIXSM4TKFTQSF/action/author_attestation","sign_citation":"https://pith.science/pith/2ILTPYNY3GF4AEIXSM4TKFTQSF/action/citation_signature","submit_replication":"https://pith.science/pith/2ILTPYNY3GF4AEIXSM4TKFTQSF/action/replication_record"}},"created_at":"2026-07-05T02:25:15.034034+00:00","updated_at":"2026-07-05T02:25:15.034034+00:00"}