{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:KNTEXXIJNY5X3D6AUPM64GHEUG","short_pith_number":"pith:KNTEXXIJ","schema_version":"1.0","canonical_sha256":"53664bdd096e3b7d8fc0a3d9ee18e4a1911daf28c8c5c179f7fd7e749cb6c49d","source":{"kind":"arxiv","id":"2002.03155","version":3},"attestation_state":"computed","paper":{"title":"Random Features Strengthen Graph Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Hisashi Kashima, Makoto Yamada, Ryoma Sato","submitted_at":"2020-02-08T12:47:29Z","abstract_excerpt":"Graph neural networks (GNNs) are powerful machine learning models for various graph learning tasks. Recently, the limitations of the expressive power of various GNN models have been revealed. For example, GNNs cannot distinguish some non-isomorphic graphs and they cannot learn efficient graph algorithms. In this paper, we demonstrate that GNNs become powerful just by adding a random feature to each node. We prove that the random features enable GNNs to learn almost optimal polynomial-time approximation algorithms for the minimum dominating set problem and maximum matching problem in terms of a"},"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":"2002.03155","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-02-08T12:47:29Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"08dee325a510df570b556ff2235e79b31157193d7449e244cc50b45ea06fdea6","abstract_canon_sha256":"4ac481fb193a931e1a9b306613c46d6c9c7eb80b21a05414a3b5f9734a899fd2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:07:27.315278Z","signature_b64":"OWY+Hq9n7ZYSszi+ZjrcIJ0cT+dKWtcxR+5na1h6HvDrsHcC95PxWBCiabe7FYbg9b+DKRXveFXi6KVWOy90Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"53664bdd096e3b7d8fc0a3d9ee18e4a1911daf28c8c5c179f7fd7e749cb6c49d","last_reissued_at":"2026-07-05T02:07:27.314798Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:07:27.314798Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Random Features Strengthen Graph Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Hisashi Kashima, Makoto Yamada, Ryoma Sato","submitted_at":"2020-02-08T12:47:29Z","abstract_excerpt":"Graph neural networks (GNNs) are powerful machine learning models for various graph learning tasks. Recently, the limitations of the expressive power of various GNN models have been revealed. For example, GNNs cannot distinguish some non-isomorphic graphs and they cannot learn efficient graph algorithms. In this paper, we demonstrate that GNNs become powerful just by adding a random feature to each node. We prove that the random features enable GNNs to learn almost optimal polynomial-time approximation algorithms for the minimum dominating set problem and maximum matching problem in terms of a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2002.03155","kind":"arxiv","version":3},"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/2002.03155/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":"2002.03155","created_at":"2026-07-05T02:07:27.314861+00:00"},{"alias_kind":"arxiv_version","alias_value":"2002.03155v3","created_at":"2026-07-05T02:07:27.314861+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2002.03155","created_at":"2026-07-05T02:07:27.314861+00:00"},{"alias_kind":"pith_short_12","alias_value":"KNTEXXIJNY5X","created_at":"2026-07-05T02:07:27.314861+00:00"},{"alias_kind":"pith_short_16","alias_value":"KNTEXXIJNY5X3D6A","created_at":"2026-07-05T02:07:27.314861+00:00"},{"alias_kind":"pith_short_8","alias_value":"KNTEXXIJ","created_at":"2026-07-05T02:07:27.314861+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2104.13478","citing_title":"Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges","ref_index":76,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KNTEXXIJNY5X3D6AUPM64GHEUG","json":"https://pith.science/pith/KNTEXXIJNY5X3D6AUPM64GHEUG.json","graph_json":"https://pith.science/api/pith-number/KNTEXXIJNY5X3D6AUPM64GHEUG/graph.json","events_json":"https://pith.science/api/pith-number/KNTEXXIJNY5X3D6AUPM64GHEUG/events.json","paper":"https://pith.science/paper/KNTEXXIJ"},"agent_actions":{"view_html":"https://pith.science/pith/KNTEXXIJNY5X3D6AUPM64GHEUG","download_json":"https://pith.science/pith/KNTEXXIJNY5X3D6AUPM64GHEUG.json","view_paper":"https://pith.science/paper/KNTEXXIJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2002.03155&json=true","fetch_graph":"https://pith.science/api/pith-number/KNTEXXIJNY5X3D6AUPM64GHEUG/graph.json","fetch_events":"https://pith.science/api/pith-number/KNTEXXIJNY5X3D6AUPM64GHEUG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KNTEXXIJNY5X3D6AUPM64GHEUG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KNTEXXIJNY5X3D6AUPM64GHEUG/action/storage_attestation","attest_author":"https://pith.science/pith/KNTEXXIJNY5X3D6AUPM64GHEUG/action/author_attestation","sign_citation":"https://pith.science/pith/KNTEXXIJNY5X3D6AUPM64GHEUG/action/citation_signature","submit_replication":"https://pith.science/pith/KNTEXXIJNY5X3D6AUPM64GHEUG/action/replication_record"}},"created_at":"2026-07-05T02:07:27.314861+00:00","updated_at":"2026-07-05T02:07:27.314861+00:00"}