{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:TZXCTL2EVMYNGD7DMBYZRDOV22","short_pith_number":"pith:TZXCTL2E","schema_version":"1.0","canonical_sha256":"9e6e29af44ab30d30fe36071988dd5d6aec6f3b1ac84bafae9bbbe47260dc6ee","source":{"kind":"arxiv","id":"1812.09624","version":1},"attestation_state":"computed","paper":{"title":"Can VAEs Generate Novel Examples?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Alican Bozkurt, Babak Esmaeili, Dana H. Brooks, Jan-Willem van de Meent, Jennifer G. Dy","submitted_at":"2018-12-22T23:43:42Z","abstract_excerpt":"An implicit goal in works on deep generative models is that such models should be able to generate novel examples that were not previously seen in the training data. In this paper, we investigate to what extent this property holds for widely employed variational autoencoder (VAE) architectures. VAEs maximize a lower bound on the log marginal likelihood, which implies that they will in principle overfit the training data when provided with a sufficiently expressive decoder. In the limit of an infinite capacity decoder, the optimal generative model is a uniform mixture over the training data. Mo"},"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":"1812.09624","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-12-22T23:43:42Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"5a32253db0660b41188cb7e10fdfbc48678d6be7d4689758ce9382f405d948a4","abstract_canon_sha256":"6f7acff7a20d0db01af93ed2a3f039f0fdfcd395ad050df87ea83904f81238e1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:57:27.347283Z","signature_b64":"cQ2+FKH5eiVUZMu7AenX3PayM3H8tuE4+3840nmR0PZHzUIoJSZ6dA9x0CKiLffSlNUnkQaLoYs3zoADfV2GBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9e6e29af44ab30d30fe36071988dd5d6aec6f3b1ac84bafae9bbbe47260dc6ee","last_reissued_at":"2026-05-17T23:57:27.346748Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:57:27.346748Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Can VAEs Generate Novel Examples?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Alican Bozkurt, Babak Esmaeili, Dana H. Brooks, Jan-Willem van de Meent, Jennifer G. Dy","submitted_at":"2018-12-22T23:43:42Z","abstract_excerpt":"An implicit goal in works on deep generative models is that such models should be able to generate novel examples that were not previously seen in the training data. In this paper, we investigate to what extent this property holds for widely employed variational autoencoder (VAE) architectures. VAEs maximize a lower bound on the log marginal likelihood, which implies that they will in principle overfit the training data when provided with a sufficiently expressive decoder. In the limit of an infinite capacity decoder, the optimal generative model is a uniform mixture over the training data. Mo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1812.09624","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":"1812.09624","created_at":"2026-05-17T23:57:27.346821+00:00"},{"alias_kind":"arxiv_version","alias_value":"1812.09624v1","created_at":"2026-05-17T23:57:27.346821+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1812.09624","created_at":"2026-05-17T23:57:27.346821+00:00"},{"alias_kind":"pith_short_12","alias_value":"TZXCTL2EVMYN","created_at":"2026-05-18T12:32:56.356000+00:00"},{"alias_kind":"pith_short_16","alias_value":"TZXCTL2EVMYNGD7D","created_at":"2026-05-18T12:32:56.356000+00:00"},{"alias_kind":"pith_short_8","alias_value":"TZXCTL2E","created_at":"2026-05-18T12:32:56.356000+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/TZXCTL2EVMYNGD7DMBYZRDOV22","json":"https://pith.science/pith/TZXCTL2EVMYNGD7DMBYZRDOV22.json","graph_json":"https://pith.science/api/pith-number/TZXCTL2EVMYNGD7DMBYZRDOV22/graph.json","events_json":"https://pith.science/api/pith-number/TZXCTL2EVMYNGD7DMBYZRDOV22/events.json","paper":"https://pith.science/paper/TZXCTL2E"},"agent_actions":{"view_html":"https://pith.science/pith/TZXCTL2EVMYNGD7DMBYZRDOV22","download_json":"https://pith.science/pith/TZXCTL2EVMYNGD7DMBYZRDOV22.json","view_paper":"https://pith.science/paper/TZXCTL2E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1812.09624&json=true","fetch_graph":"https://pith.science/api/pith-number/TZXCTL2EVMYNGD7DMBYZRDOV22/graph.json","fetch_events":"https://pith.science/api/pith-number/TZXCTL2EVMYNGD7DMBYZRDOV22/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TZXCTL2EVMYNGD7DMBYZRDOV22/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TZXCTL2EVMYNGD7DMBYZRDOV22/action/storage_attestation","attest_author":"https://pith.science/pith/TZXCTL2EVMYNGD7DMBYZRDOV22/action/author_attestation","sign_citation":"https://pith.science/pith/TZXCTL2EVMYNGD7DMBYZRDOV22/action/citation_signature","submit_replication":"https://pith.science/pith/TZXCTL2EVMYNGD7DMBYZRDOV22/action/replication_record"}},"created_at":"2026-05-17T23:57:27.346821+00:00","updated_at":"2026-05-17T23:57:27.346821+00:00"}