{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2016:VXJR6KZ5MHPFYI4NYUYMXJBBUS","short_pith_number":"pith:VXJR6KZ5","schema_version":"1.0","canonical_sha256":"add31f2b3d61de5c238dc530cba421a4811df04013206b06424c54a1ce4d590c","source":{"kind":"arxiv","id":"1609.02200","version":2},"attestation_state":"computed","paper":{"title":"Discrete Variational Autoencoders","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Jason Tyler Rolfe","submitted_at":"2016-09-07T21:41:32Z","abstract_excerpt":"Probabilistic models with discrete latent variables naturally capture datasets composed of discrete classes. However, they are difficult to train efficiently, since backpropagation through discrete variables is generally not possible. We present a novel method to train a class of probabilistic models with discrete latent variables using the variational autoencoder framework, including backpropagation through the discrete latent variables. The associated class of probabilistic models comprises an undirected discrete component and a directed hierarchical continuous component. The discrete compon"},"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":"1609.02200","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2016-09-07T21:41:32Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"e06196f8ccec1f0c2b76aa8a6bd60b6568cd48f33b1ccac38c17c014d5538f7a","abstract_canon_sha256":"16cdca13beeb429f4d0ef3508490c5210e03aa547306e64577264eb2e755413c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:45:58.625372Z","signature_b64":"7IwbeD+bAPteScj+kpLLcUfexKWYSKDFMypPxg0I16vLIosi9m7+7mk+5bP3E6HHkJHUoWs2e9fArq5aajn0DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"add31f2b3d61de5c238dc530cba421a4811df04013206b06424c54a1ce4d590c","last_reissued_at":"2026-05-18T00:45:58.624936Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:45:58.624936Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Discrete Variational Autoencoders","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Jason Tyler Rolfe","submitted_at":"2016-09-07T21:41:32Z","abstract_excerpt":"Probabilistic models with discrete latent variables naturally capture datasets composed of discrete classes. However, they are difficult to train efficiently, since backpropagation through discrete variables is generally not possible. We present a novel method to train a class of probabilistic models with discrete latent variables using the variational autoencoder framework, including backpropagation through the discrete latent variables. The associated class of probabilistic models comprises an undirected discrete component and a directed hierarchical continuous component. The discrete compon"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1609.02200","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":""},"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":"1609.02200","created_at":"2026-05-18T00:45:58.624999+00:00"},{"alias_kind":"arxiv_version","alias_value":"1609.02200v2","created_at":"2026-05-18T00:45:58.624999+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1609.02200","created_at":"2026-05-18T00:45:58.624999+00:00"},{"alias_kind":"pith_short_12","alias_value":"VXJR6KZ5MHPF","created_at":"2026-05-18T12:30:48.956258+00:00"},{"alias_kind":"pith_short_16","alias_value":"VXJR6KZ5MHPFYI4N","created_at":"2026-05-18T12:30:48.956258+00:00"},{"alias_kind":"pith_short_8","alias_value":"VXJR6KZ5","created_at":"2026-05-18T12:30:48.956258+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":7,"sample":[{"citing_arxiv_id":"2607.07840","citing_title":"GradInf: Gradient Estimation as Probabilistic Inference","ref_index":88,"is_internal_anchor":true},{"citing_arxiv_id":"2606.04432","citing_title":"DSA: Dynamic Step Allocation for Fast Autoregressive Video Generation","ref_index":46,"is_internal_anchor":true},{"citing_arxiv_id":"2605.17131","citing_title":"A Survey on Deep Learning Architectures for Point Cloud Classification and Segmentation","ref_index":78,"is_internal_anchor":true},{"citing_arxiv_id":"2604.00919","citing_title":"Multi-Mode Quantum Annealing for Generative Representation Learning with Boltzmann Priors","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2605.17131","citing_title":"A Survey on Deep Learning Architectures for Point Cloud Classification and Segmentation","ref_index":78,"is_internal_anchor":true},{"citing_arxiv_id":"2602.15451","citing_title":"Molecular Design beyond Training Data with Novel Extended Objective Functionals of Generative AI Models Driven by Quantum Annealing Computer","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14297","citing_title":"Policy Optimization in Hybrid Discrete-Continuous Action Spaces via Mixed Gradients","ref_index":104,"is_internal_anchor":true},{"citing_arxiv_id":"2206.10789","citing_title":"Scaling Autoregressive Models for Content-Rich Text-to-Image Generation","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VXJR6KZ5MHPFYI4NYUYMXJBBUS","json":"https://pith.science/pith/VXJR6KZ5MHPFYI4NYUYMXJBBUS.json","graph_json":"https://pith.science/api/pith-number/VXJR6KZ5MHPFYI4NYUYMXJBBUS/graph.json","events_json":"https://pith.science/api/pith-number/VXJR6KZ5MHPFYI4NYUYMXJBBUS/events.json","paper":"https://pith.science/paper/VXJR6KZ5"},"agent_actions":{"view_html":"https://pith.science/pith/VXJR6KZ5MHPFYI4NYUYMXJBBUS","download_json":"https://pith.science/pith/VXJR6KZ5MHPFYI4NYUYMXJBBUS.json","view_paper":"https://pith.science/paper/VXJR6KZ5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1609.02200&json=true","fetch_graph":"https://pith.science/api/pith-number/VXJR6KZ5MHPFYI4NYUYMXJBBUS/graph.json","fetch_events":"https://pith.science/api/pith-number/VXJR6KZ5MHPFYI4NYUYMXJBBUS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VXJR6KZ5MHPFYI4NYUYMXJBBUS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VXJR6KZ5MHPFYI4NYUYMXJBBUS/action/storage_attestation","attest_author":"https://pith.science/pith/VXJR6KZ5MHPFYI4NYUYMXJBBUS/action/author_attestation","sign_citation":"https://pith.science/pith/VXJR6KZ5MHPFYI4NYUYMXJBBUS/action/citation_signature","submit_replication":"https://pith.science/pith/VXJR6KZ5MHPFYI4NYUYMXJBBUS/action/replication_record"}},"created_at":"2026-05-18T00:45:58.624999+00:00","updated_at":"2026-05-18T00:45:58.624999+00:00"}