{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:UUOA3Z5ZH5UG2LETKKA65SJPK3","short_pith_number":"pith:UUOA3Z5Z","schema_version":"1.0","canonical_sha256":"a51c0de7b93f686d2c935281eec92f56dfaf8157b48c528972d2a21749148448","source":{"kind":"arxiv","id":"1910.11831","version":5},"attestation_state":"computed","paper":{"title":"Stabilizing DARTS with Amended Gradient Estimation on Architectural Parameters","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Changping Hu, Kaifeng Bi, Lingxi Xie, Longhui Wei, Qi Tian, Xin Chen","submitted_at":"2019-10-25T16:31:25Z","abstract_excerpt":"DARTS is a popular algorithm for neural architecture search (NAS). Despite its great advantage in search efficiency, DARTS often suffers weak stability, which reflects in the large variation among individual trials as well as the sensitivity to the hyper-parameters of the search process. This paper owes such instability to an optimization gap between the super-network and its sub-networks, namely, improving the validation accuracy of the super-network does not necessarily lead to a higher expectation on the performance of the sampled sub-networks. Then, we point out that the gap is due to the "},"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":"1910.11831","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-10-25T16:31:25Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"68b43b3add4de4b0be5b2577753d2d5edf021b907c5605000bf0add940415815","abstract_canon_sha256":"3aabb4b908eab2f1df0fcb27f2946135796cef0ca91950cee62723fcc6afa691"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:59:54.796199Z","signature_b64":"6dFEi4uWLGg08668QSZz5FLx6GSZQSfkM3k6qiej+gMKDd7cCJgmq/2Mi6sG81TLssShZqYCRmm1bubD+T/eAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a51c0de7b93f686d2c935281eec92f56dfaf8157b48c528972d2a21749148448","last_reissued_at":"2026-07-05T00:59:54.795708Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:59:54.795708Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Stabilizing DARTS with Amended Gradient Estimation on Architectural Parameters","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Changping Hu, Kaifeng Bi, Lingxi Xie, Longhui Wei, Qi Tian, Xin Chen","submitted_at":"2019-10-25T16:31:25Z","abstract_excerpt":"DARTS is a popular algorithm for neural architecture search (NAS). Despite its great advantage in search efficiency, DARTS often suffers weak stability, which reflects in the large variation among individual trials as well as the sensitivity to the hyper-parameters of the search process. This paper owes such instability to an optimization gap between the super-network and its sub-networks, namely, improving the validation accuracy of the super-network does not necessarily lead to a higher expectation on the performance of the sampled sub-networks. Then, we point out that the gap is due to the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1910.11831","kind":"arxiv","version":5},"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/1910.11831/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":"1910.11831","created_at":"2026-07-05T00:59:54.795776+00:00"},{"alias_kind":"arxiv_version","alias_value":"1910.11831v5","created_at":"2026-07-05T00:59:54.795776+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1910.11831","created_at":"2026-07-05T00:59:54.795776+00:00"},{"alias_kind":"pith_short_12","alias_value":"UUOA3Z5ZH5UG","created_at":"2026-07-05T00:59:54.795776+00:00"},{"alias_kind":"pith_short_16","alias_value":"UUOA3Z5ZH5UG2LET","created_at":"2026-07-05T00:59:54.795776+00:00"},{"alias_kind":"pith_short_8","alias_value":"UUOA3Z5Z","created_at":"2026-07-05T00:59:54.795776+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.16533","citing_title":"confopt: A Library for Implementation and Evaluation of Gradient-based One-Shot NAS Methods","ref_index":2,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UUOA3Z5ZH5UG2LETKKA65SJPK3","json":"https://pith.science/pith/UUOA3Z5ZH5UG2LETKKA65SJPK3.json","graph_json":"https://pith.science/api/pith-number/UUOA3Z5ZH5UG2LETKKA65SJPK3/graph.json","events_json":"https://pith.science/api/pith-number/UUOA3Z5ZH5UG2LETKKA65SJPK3/events.json","paper":"https://pith.science/paper/UUOA3Z5Z"},"agent_actions":{"view_html":"https://pith.science/pith/UUOA3Z5ZH5UG2LETKKA65SJPK3","download_json":"https://pith.science/pith/UUOA3Z5ZH5UG2LETKKA65SJPK3.json","view_paper":"https://pith.science/paper/UUOA3Z5Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1910.11831&json=true","fetch_graph":"https://pith.science/api/pith-number/UUOA3Z5ZH5UG2LETKKA65SJPK3/graph.json","fetch_events":"https://pith.science/api/pith-number/UUOA3Z5ZH5UG2LETKKA65SJPK3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UUOA3Z5ZH5UG2LETKKA65SJPK3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UUOA3Z5ZH5UG2LETKKA65SJPK3/action/storage_attestation","attest_author":"https://pith.science/pith/UUOA3Z5ZH5UG2LETKKA65SJPK3/action/author_attestation","sign_citation":"https://pith.science/pith/UUOA3Z5ZH5UG2LETKKA65SJPK3/action/citation_signature","submit_replication":"https://pith.science/pith/UUOA3Z5ZH5UG2LETKKA65SJPK3/action/replication_record"}},"created_at":"2026-07-05T00:59:54.795776+00:00","updated_at":"2026-07-05T00:59:54.795776+00:00"}