{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:ZH772F2AHUBG4EBT7G73BEWDBG","short_pith_number":"pith:ZH772F2A","schema_version":"1.0","canonical_sha256":"c9fffd17403d026e1033f9bfb092c3099f093a8d23807a542c40a0e1719250ea","source":{"kind":"arxiv","id":"1801.02144","version":1},"attestation_state":"computed","paper":{"title":"Covariant Compositional Networks For Learning Graphs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Brandon Anderson, Horace Pan, Hy Truong Son, Risi Kondor, Shubhendu Trivedi","submitted_at":"2018-01-07T06:14:55Z","abstract_excerpt":"Most existing neural networks for learning graphs address permutation invariance by conceiving of the network as a message passing scheme, where each node sums the feature vectors coming from its neighbors. We argue that this imposes a limitation on their representation power, and instead propose a new general architecture for representing objects consisting of a hierarchy of parts, which we call Covariant Compositional Networks (CCNs). Here, covariance means that the activation of each neuron must transform in a specific way under permutations, similarly to steerability in CNNs. We achieve co"},"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":"1801.02144","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-01-07T06:14:55Z","cross_cats_sorted":[],"title_canon_sha256":"13221c0ea8c081510d9674f6bdc39dd272a4e8368686af9fe30f6ed9b0f8af26","abstract_canon_sha256":"50c1879c44e5405e71c59645fbab928b612ef4ec5662ed907782c44e2d150fd9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:26:34.723472Z","signature_b64":"XLg7ei6AEVFFojaBXpaTHccqs8KSKxrIiJOpIxZCK/C6n+v9WQ7cLOUFeFA3npkvWZFgy3knzAoW4Fcunfd1Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c9fffd17403d026e1033f9bfb092c3099f093a8d23807a542c40a0e1719250ea","last_reissued_at":"2026-05-18T00:26:34.722758Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:26:34.722758Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Covariant Compositional Networks For Learning Graphs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Brandon Anderson, Horace Pan, Hy Truong Son, Risi Kondor, Shubhendu Trivedi","submitted_at":"2018-01-07T06:14:55Z","abstract_excerpt":"Most existing neural networks for learning graphs address permutation invariance by conceiving of the network as a message passing scheme, where each node sums the feature vectors coming from its neighbors. We argue that this imposes a limitation on their representation power, and instead propose a new general architecture for representing objects consisting of a hierarchy of parts, which we call Covariant Compositional Networks (CCNs). Here, covariance means that the activation of each neuron must transform in a specific way under permutations, similarly to steerability in CNNs. We achieve co"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1801.02144","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":""},"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":"1801.02144","created_at":"2026-05-18T00:26:34.722868+00:00"},{"alias_kind":"arxiv_version","alias_value":"1801.02144v1","created_at":"2026-05-18T00:26:34.722868+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1801.02144","created_at":"2026-05-18T00:26:34.722868+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZH772F2AHUBG","created_at":"2026-05-18T12:33:07.085635+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZH772F2AHUBG4EBT","created_at":"2026-05-18T12:33:07.085635+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZH772F2A","created_at":"2026-05-18T12:33:07.085635+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"1806.01261","citing_title":"Relational inductive biases, deep learning, and graph networks","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZH772F2AHUBG4EBT7G73BEWDBG","json":"https://pith.science/pith/ZH772F2AHUBG4EBT7G73BEWDBG.json","graph_json":"https://pith.science/api/pith-number/ZH772F2AHUBG4EBT7G73BEWDBG/graph.json","events_json":"https://pith.science/api/pith-number/ZH772F2AHUBG4EBT7G73BEWDBG/events.json","paper":"https://pith.science/paper/ZH772F2A"},"agent_actions":{"view_html":"https://pith.science/pith/ZH772F2AHUBG4EBT7G73BEWDBG","download_json":"https://pith.science/pith/ZH772F2AHUBG4EBT7G73BEWDBG.json","view_paper":"https://pith.science/paper/ZH772F2A","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1801.02144&json=true","fetch_graph":"https://pith.science/api/pith-number/ZH772F2AHUBG4EBT7G73BEWDBG/graph.json","fetch_events":"https://pith.science/api/pith-number/ZH772F2AHUBG4EBT7G73BEWDBG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZH772F2AHUBG4EBT7G73BEWDBG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZH772F2AHUBG4EBT7G73BEWDBG/action/storage_attestation","attest_author":"https://pith.science/pith/ZH772F2AHUBG4EBT7G73BEWDBG/action/author_attestation","sign_citation":"https://pith.science/pith/ZH772F2AHUBG4EBT7G73BEWDBG/action/citation_signature","submit_replication":"https://pith.science/pith/ZH772F2AHUBG4EBT7G73BEWDBG/action/replication_record"}},"created_at":"2026-05-18T00:26:34.722868+00:00","updated_at":"2026-05-18T00:26:34.722868+00:00"}