{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:6W4FI2OBVN2F2G6KUDFVQCFRZG","short_pith_number":"pith:6W4FI2OB","schema_version":"1.0","canonical_sha256":"f5b85469c1ab745d1bcaa0cb5808b1c98a92f79407409403d04c097a4a4f32f4","source":{"kind":"arxiv","id":"1707.03389","version":3},"attestation_state":"computed","paper":{"title":"SCAN: Learning Hierarchical Compositional Visual Concepts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Alexander Lerchner, Arka Pal, Christopher P Burgess, Demis Hassabis, Irina Higgins, Loic Matthey, Matko Bosnjak, Matthew Botvinick, Murray Shanahan, Nicolas Sonnerat","submitted_at":"2017-07-11T17:58:45Z","abstract_excerpt":"The seemingly infinite diversity of the natural world arises from a relatively small set of coherent rules, such as the laws of physics or chemistry. We conjecture that these rules give rise to regularities that can be discovered through primarily unsupervised experiences and represented as abstract concepts. If such representations are compositional and hierarchical, they can be recombined into an exponentially large set of new concepts. This paper describes SCAN (Symbol-Concept Association Network), a new framework for learning such abstractions in the visual domain. SCAN learns concepts thr"},"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":"1707.03389","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2017-07-11T17:58:45Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"73096641e0df9e3182e74e7b95810bfe4b11c19806cf72faeab323ecfda75019","abstract_canon_sha256":"c21ec8881ffd024f4e3dacf004efa33d2927eb97a05afe1164d99a8bcf61f50a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:14:06.775620Z","signature_b64":"was0n06KP4Q+s5hnrV9mpagaSH8ejNatVwLg4twdqttia34A6n07QeaSKcuUnP8wNHPABLCRibo0NNzdWk7MCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f5b85469c1ab745d1bcaa0cb5808b1c98a92f79407409403d04c097a4a4f32f4","last_reissued_at":"2026-05-18T00:14:06.775151Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:14:06.775151Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SCAN: Learning Hierarchical Compositional Visual Concepts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Alexander Lerchner, Arka Pal, Christopher P Burgess, Demis Hassabis, Irina Higgins, Loic Matthey, Matko Bosnjak, Matthew Botvinick, Murray Shanahan, Nicolas Sonnerat","submitted_at":"2017-07-11T17:58:45Z","abstract_excerpt":"The seemingly infinite diversity of the natural world arises from a relatively small set of coherent rules, such as the laws of physics or chemistry. We conjecture that these rules give rise to regularities that can be discovered through primarily unsupervised experiences and represented as abstract concepts. If such representations are compositional and hierarchical, they can be recombined into an exponentially large set of new concepts. This paper describes SCAN (Symbol-Concept Association Network), a new framework for learning such abstractions in the visual domain. SCAN learns concepts thr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1707.03389","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":""},"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":"1707.03389","created_at":"2026-05-18T00:14:06.775226+00:00"},{"alias_kind":"arxiv_version","alias_value":"1707.03389v3","created_at":"2026-05-18T00:14:06.775226+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1707.03389","created_at":"2026-05-18T00:14:06.775226+00:00"},{"alias_kind":"pith_short_12","alias_value":"6W4FI2OBVN2F","created_at":"2026-05-18T12:31:03.183658+00:00"},{"alias_kind":"pith_short_16","alias_value":"6W4FI2OBVN2F2G6K","created_at":"2026-05-18T12:31:03.183658+00:00"},{"alias_kind":"pith_short_8","alias_value":"6W4FI2OB","created_at":"2026-05-18T12:31:03.183658+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.20732","citing_title":"Deep Attention Reweighting: Post-Hoc Attention-Based Feature Aggregation in CNNs for Disentangling Core and Spurious Features under Spurious Correlations","ref_index":19,"is_internal_anchor":true},{"citing_arxiv_id":"2605.03413","citing_title":"Learning to Theorize the World from Observation","ref_index":67,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6W4FI2OBVN2F2G6KUDFVQCFRZG","json":"https://pith.science/pith/6W4FI2OBVN2F2G6KUDFVQCFRZG.json","graph_json":"https://pith.science/api/pith-number/6W4FI2OBVN2F2G6KUDFVQCFRZG/graph.json","events_json":"https://pith.science/api/pith-number/6W4FI2OBVN2F2G6KUDFVQCFRZG/events.json","paper":"https://pith.science/paper/6W4FI2OB"},"agent_actions":{"view_html":"https://pith.science/pith/6W4FI2OBVN2F2G6KUDFVQCFRZG","download_json":"https://pith.science/pith/6W4FI2OBVN2F2G6KUDFVQCFRZG.json","view_paper":"https://pith.science/paper/6W4FI2OB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1707.03389&json=true","fetch_graph":"https://pith.science/api/pith-number/6W4FI2OBVN2F2G6KUDFVQCFRZG/graph.json","fetch_events":"https://pith.science/api/pith-number/6W4FI2OBVN2F2G6KUDFVQCFRZG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6W4FI2OBVN2F2G6KUDFVQCFRZG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6W4FI2OBVN2F2G6KUDFVQCFRZG/action/storage_attestation","attest_author":"https://pith.science/pith/6W4FI2OBVN2F2G6KUDFVQCFRZG/action/author_attestation","sign_citation":"https://pith.science/pith/6W4FI2OBVN2F2G6KUDFVQCFRZG/action/citation_signature","submit_replication":"https://pith.science/pith/6W4FI2OBVN2F2G6KUDFVQCFRZG/action/replication_record"}},"created_at":"2026-05-18T00:14:06.775226+00:00","updated_at":"2026-05-18T00:14:06.775226+00:00"}