{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:LGRWX2EJNWXLKLU4JU7CXALC5E","short_pith_number":"pith:LGRWX2EJ","schema_version":"1.0","canonical_sha256":"59a36be8896daeb52e9c4d3e2b8162e921586ff1052cf1ea04ae85a822578ed7","source":{"kind":"arxiv","id":"2107.05680","version":2},"attestation_state":"computed","paper":{"title":"Hidden Convexity of Wasserstein GANs: Interpretable Generative Models with Closed-Form Solutions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","eess.IV","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Arda Sahiner, Batu Ozturkler, Burak Bartan, John Pauly, Mert Pilanci, Morteza Mardani, Tolga Ergen","submitted_at":"2021-07-12T18:33:49Z","abstract_excerpt":"Generative Adversarial Networks (GANs) are commonly used for modeling complex distributions of data. Both the generators and discriminators of GANs are often modeled by neural networks, posing a non-transparent optimization problem which is non-convex and non-concave over the generator and discriminator, respectively. Such networks are often heuristically optimized with gradient descent-ascent (GDA), but it is unclear whether the optimization problem contains any saddle points, or whether heuristic methods can find them in practice. In this work, we analyze the training of Wasserstein GANs wit"},"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":"2107.05680","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-07-12T18:33:49Z","cross_cats_sorted":["cs.CV","eess.IV","math.OC","stat.ML"],"title_canon_sha256":"ab9dde282b51eabde44061d6cef4469eb96586e7c4bc242c85037c790558dfb5","abstract_canon_sha256":"9c98900f442e5907e0307afdf82bccb860e1931ab55553c1e432a3ce677025ae"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:06:43.915009Z","signature_b64":"Qgi5QkiRY/ztLRPu/azKvaAYCui9Vy0eS5avuGgCEOCZK5l+VUCQzvRBMKInRvklY4FRMxnS9k+U8PBYcvh2AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"59a36be8896daeb52e9c4d3e2b8162e921586ff1052cf1ea04ae85a822578ed7","last_reissued_at":"2026-07-05T04:06:43.914445Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:06:43.914445Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Hidden Convexity of Wasserstein GANs: Interpretable Generative Models with Closed-Form Solutions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","eess.IV","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Arda Sahiner, Batu Ozturkler, Burak Bartan, John Pauly, Mert Pilanci, Morteza Mardani, Tolga Ergen","submitted_at":"2021-07-12T18:33:49Z","abstract_excerpt":"Generative Adversarial Networks (GANs) are commonly used for modeling complex distributions of data. Both the generators and discriminators of GANs are often modeled by neural networks, posing a non-transparent optimization problem which is non-convex and non-concave over the generator and discriminator, respectively. Such networks are often heuristically optimized with gradient descent-ascent (GDA), but it is unclear whether the optimization problem contains any saddle points, or whether heuristic methods can find them in practice. In this work, we analyze the training of Wasserstein GANs wit"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.05680","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2107.05680/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":"2107.05680","created_at":"2026-07-05T04:06:43.914506+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.05680v2","created_at":"2026-07-05T04:06:43.914506+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.05680","created_at":"2026-07-05T04:06:43.914506+00:00"},{"alias_kind":"pith_short_12","alias_value":"LGRWX2EJNWXL","created_at":"2026-07-05T04:06:43.914506+00:00"},{"alias_kind":"pith_short_16","alias_value":"LGRWX2EJNWXLKLU4","created_at":"2026-07-05T04:06:43.914506+00:00"},{"alias_kind":"pith_short_8","alias_value":"LGRWX2EJ","created_at":"2026-07-05T04:06:43.914506+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.04260","citing_title":"Realistic Image-to-Image Machine Unlearning via Decoupling and Knowledge Retention","ref_index":27,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LGRWX2EJNWXLKLU4JU7CXALC5E","json":"https://pith.science/pith/LGRWX2EJNWXLKLU4JU7CXALC5E.json","graph_json":"https://pith.science/api/pith-number/LGRWX2EJNWXLKLU4JU7CXALC5E/graph.json","events_json":"https://pith.science/api/pith-number/LGRWX2EJNWXLKLU4JU7CXALC5E/events.json","paper":"https://pith.science/paper/LGRWX2EJ"},"agent_actions":{"view_html":"https://pith.science/pith/LGRWX2EJNWXLKLU4JU7CXALC5E","download_json":"https://pith.science/pith/LGRWX2EJNWXLKLU4JU7CXALC5E.json","view_paper":"https://pith.science/paper/LGRWX2EJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.05680&json=true","fetch_graph":"https://pith.science/api/pith-number/LGRWX2EJNWXLKLU4JU7CXALC5E/graph.json","fetch_events":"https://pith.science/api/pith-number/LGRWX2EJNWXLKLU4JU7CXALC5E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LGRWX2EJNWXLKLU4JU7CXALC5E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LGRWX2EJNWXLKLU4JU7CXALC5E/action/storage_attestation","attest_author":"https://pith.science/pith/LGRWX2EJNWXLKLU4JU7CXALC5E/action/author_attestation","sign_citation":"https://pith.science/pith/LGRWX2EJNWXLKLU4JU7CXALC5E/action/citation_signature","submit_replication":"https://pith.science/pith/LGRWX2EJNWXLKLU4JU7CXALC5E/action/replication_record"}},"created_at":"2026-07-05T04:06:43.914506+00:00","updated_at":"2026-07-05T04:06:43.914506+00:00"}