{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:G75B5ZPVZ6YVA322RMLDBW3VVG","short_pith_number":"pith:G75B5ZPV","schema_version":"1.0","canonical_sha256":"37fa1ee5f5cfb1506f5a8b1630db75a9b2ced77c89471128e23f94fe9e778789","source":{"kind":"arxiv","id":"1801.04406","version":4},"attestation_state":"computed","paper":{"title":"Which Training Methods for GANs do actually Converge?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.GT"],"primary_cat":"cs.LG","authors_text":"Andreas Geiger, Lars Mescheder, Sebastian Nowozin","submitted_at":"2018-01-13T09:42:26Z","abstract_excerpt":"Recent work has shown local convergence of GAN training for absolutely continuous data and generator distributions. In this paper, we show that the requirement of absolute continuity is necessary: we describe a simple yet prototypical counterexample showing that in the more realistic case of distributions that are not absolutely continuous, unregularized GAN training is not always convergent. Furthermore, we discuss regularization strategies that were recently proposed to stabilize GAN training. Our analysis shows that GAN training with instance noise or zero-centered gradient penalties conver"},"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":"1801.04406","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-01-13T09:42:26Z","cross_cats_sorted":["cs.AI","cs.GT"],"title_canon_sha256":"bd7369cc097303d42b7b79ba0f623ca21f9fb2c587ca600b62926da7cba66115","abstract_canon_sha256":"3f0dd5cb55b41b80d087d20bd9cab5e8bf036ebb80b3d988ba300ab2d1e17510"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:09:27.492037Z","signature_b64":"TjeVtz+Q8SW4KRMnlwwFjsJa2pnnOFTfWuGUkOsdWx8wkx6AtKf0nQBjqp7AT7oDp5ra5a9SSQG/XOaS+wc7Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"37fa1ee5f5cfb1506f5a8b1630db75a9b2ced77c89471128e23f94fe9e778789","last_reissued_at":"2026-05-18T00:09:27.491621Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:09:27.491621Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Which Training Methods for GANs do actually Converge?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.GT"],"primary_cat":"cs.LG","authors_text":"Andreas Geiger, Lars Mescheder, Sebastian Nowozin","submitted_at":"2018-01-13T09:42:26Z","abstract_excerpt":"Recent work has shown local convergence of GAN training for absolutely continuous data and generator distributions. In this paper, we show that the requirement of absolute continuity is necessary: we describe a simple yet prototypical counterexample showing that in the more realistic case of distributions that are not absolutely continuous, unregularized GAN training is not always convergent. Furthermore, we discuss regularization strategies that were recently proposed to stabilize GAN training. Our analysis shows that GAN training with instance noise or zero-centered gradient penalties conver"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1801.04406","kind":"arxiv","version":4},"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":"1801.04406","created_at":"2026-05-18T00:09:27.491676+00:00"},{"alias_kind":"arxiv_version","alias_value":"1801.04406v4","created_at":"2026-05-18T00:09:27.491676+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1801.04406","created_at":"2026-05-18T00:09:27.491676+00:00"},{"alias_kind":"pith_short_12","alias_value":"G75B5ZPVZ6YV","created_at":"2026-05-18T12:32:25.280505+00:00"},{"alias_kind":"pith_short_16","alias_value":"G75B5ZPVZ6YVA322","created_at":"2026-05-18T12:32:25.280505+00:00"},{"alias_kind":"pith_short_8","alias_value":"G75B5ZPV","created_at":"2026-05-18T12:32:25.280505+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"1907.03361","citing_title":"Copula & Marginal Flows: Disentangling the Marginal from its Joint","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2112.10752","citing_title":"High-Resolution Image Synthesis with Latent Diffusion Models","ref_index":54,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20130","citing_title":"Pairing Regularization for Mitigating Many-to-One Collapse in GANs","ref_index":84,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/G75B5ZPVZ6YVA322RMLDBW3VVG","json":"https://pith.science/pith/G75B5ZPVZ6YVA322RMLDBW3VVG.json","graph_json":"https://pith.science/api/pith-number/G75B5ZPVZ6YVA322RMLDBW3VVG/graph.json","events_json":"https://pith.science/api/pith-number/G75B5ZPVZ6YVA322RMLDBW3VVG/events.json","paper":"https://pith.science/paper/G75B5ZPV"},"agent_actions":{"view_html":"https://pith.science/pith/G75B5ZPVZ6YVA322RMLDBW3VVG","download_json":"https://pith.science/pith/G75B5ZPVZ6YVA322RMLDBW3VVG.json","view_paper":"https://pith.science/paper/G75B5ZPV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1801.04406&json=true","fetch_graph":"https://pith.science/api/pith-number/G75B5ZPVZ6YVA322RMLDBW3VVG/graph.json","fetch_events":"https://pith.science/api/pith-number/G75B5ZPVZ6YVA322RMLDBW3VVG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G75B5ZPVZ6YVA322RMLDBW3VVG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G75B5ZPVZ6YVA322RMLDBW3VVG/action/storage_attestation","attest_author":"https://pith.science/pith/G75B5ZPVZ6YVA322RMLDBW3VVG/action/author_attestation","sign_citation":"https://pith.science/pith/G75B5ZPVZ6YVA322RMLDBW3VVG/action/citation_signature","submit_replication":"https://pith.science/pith/G75B5ZPVZ6YVA322RMLDBW3VVG/action/replication_record"}},"created_at":"2026-05-18T00:09:27.491676+00:00","updated_at":"2026-05-18T00:09:27.491676+00:00"}