{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:USA6BEZA4IYYVXDKYDQNTX4DSR","short_pith_number":"pith:USA6BEZA","schema_version":"1.0","canonical_sha256":"a481e09320e2318adc6ac0e0d9df83947a539c214a93e700c29c9ed0956cdf26","source":{"kind":"arxiv","id":"2210.14843","version":2},"attestation_state":"computed","paper":{"title":"TuneUp: A Simple Improved Training Strategy for Graph Neural Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"stat.ML","authors_text":"Edward W Huang, Jure Leskovec, Kaidi Cao, Karthik Subbian, Kenji Kawaguchi, Kexin Huang, Weihua Hu","submitted_at":"2022-10-26T16:42:14Z","abstract_excerpt":"Despite recent advances in Graph Neural Networks (GNNs), their training strategies remain largely under-explored. The conventional training strategy learns over all nodes in the original graph(s) equally, which can be sub-optimal as certain nodes are often more difficult to learn than others. Here we present TuneUp, a simple curriculum-based training strategy for improving the predictive performance of GNNs. TuneUp trains a GNN in two stages. In the first stage, TuneUp applies conventional training to obtain a strong base GNN. The base GNN tends to perform well on head nodes (nodes with large "},"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":"2210.14843","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2022-10-26T16:42:14Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"aa61eaeb8c94f4d656517d415710136ef91c353a9480cb3dccb9e4ad4a5e552a","abstract_canon_sha256":"ed10d70cda8aa7db90ac3832738d03ea1e69ffad3d99ad2818016b09d1c0d8b4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:44:47.209737Z","signature_b64":"7Rs9sxAVvq1xdvPsp17JlpPIAsKW6UTBSJlyysLlVO42iGuj917a4gsLQy+/qoGogKMB3JqguNCnkaBx3eYXDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a481e09320e2318adc6ac0e0d9df83947a539c214a93e700c29c9ed0956cdf26","last_reissued_at":"2026-07-05T06:44:47.209260Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:44:47.209260Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TuneUp: A Simple Improved Training Strategy for Graph Neural Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"stat.ML","authors_text":"Edward W Huang, Jure Leskovec, Kaidi Cao, Karthik Subbian, Kenji Kawaguchi, Kexin Huang, Weihua Hu","submitted_at":"2022-10-26T16:42:14Z","abstract_excerpt":"Despite recent advances in Graph Neural Networks (GNNs), their training strategies remain largely under-explored. The conventional training strategy learns over all nodes in the original graph(s) equally, which can be sub-optimal as certain nodes are often more difficult to learn than others. Here we present TuneUp, a simple curriculum-based training strategy for improving the predictive performance of GNNs. TuneUp trains a GNN in two stages. In the first stage, TuneUp applies conventional training to obtain a strong base GNN. The base GNN tends to perform well on head nodes (nodes with large "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.14843","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/2210.14843/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":"2210.14843","created_at":"2026-07-05T06:44:47.209324+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.14843v2","created_at":"2026-07-05T06:44:47.209324+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.14843","created_at":"2026-07-05T06:44:47.209324+00:00"},{"alias_kind":"pith_short_12","alias_value":"USA6BEZA4IYY","created_at":"2026-07-05T06:44:47.209324+00:00"},{"alias_kind":"pith_short_16","alias_value":"USA6BEZA4IYYVXDK","created_at":"2026-07-05T06:44:47.209324+00:00"},{"alias_kind":"pith_short_8","alias_value":"USA6BEZA","created_at":"2026-07-05T06:44:47.209324+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/USA6BEZA4IYYVXDKYDQNTX4DSR","json":"https://pith.science/pith/USA6BEZA4IYYVXDKYDQNTX4DSR.json","graph_json":"https://pith.science/api/pith-number/USA6BEZA4IYYVXDKYDQNTX4DSR/graph.json","events_json":"https://pith.science/api/pith-number/USA6BEZA4IYYVXDKYDQNTX4DSR/events.json","paper":"https://pith.science/paper/USA6BEZA"},"agent_actions":{"view_html":"https://pith.science/pith/USA6BEZA4IYYVXDKYDQNTX4DSR","download_json":"https://pith.science/pith/USA6BEZA4IYYVXDKYDQNTX4DSR.json","view_paper":"https://pith.science/paper/USA6BEZA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.14843&json=true","fetch_graph":"https://pith.science/api/pith-number/USA6BEZA4IYYVXDKYDQNTX4DSR/graph.json","fetch_events":"https://pith.science/api/pith-number/USA6BEZA4IYYVXDKYDQNTX4DSR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/USA6BEZA4IYYVXDKYDQNTX4DSR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/USA6BEZA4IYYVXDKYDQNTX4DSR/action/storage_attestation","attest_author":"https://pith.science/pith/USA6BEZA4IYYVXDKYDQNTX4DSR/action/author_attestation","sign_citation":"https://pith.science/pith/USA6BEZA4IYYVXDKYDQNTX4DSR/action/citation_signature","submit_replication":"https://pith.science/pith/USA6BEZA4IYYVXDKYDQNTX4DSR/action/replication_record"}},"created_at":"2026-07-05T06:44:47.209324+00:00","updated_at":"2026-07-05T06:44:47.209324+00:00"}