{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:3RF2KXXIDDLJFVOPY2G25AY2XH","short_pith_number":"pith:3RF2KXXI","schema_version":"1.0","canonical_sha256":"dc4ba55ee818d692d5cfc68dae831ab9dd83a9d9bab67928739c85131eae7172","source":{"kind":"arxiv","id":"1706.04156","version":3},"attestation_state":"computed","paper":{"title":"Gradient descent GAN optimization is locally stable","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"J. Zico Kolter, Vaishnavh Nagarajan","submitted_at":"2017-06-13T16:49:13Z","abstract_excerpt":"Despite the growing prominence of generative adversarial networks (GANs), optimization in GANs is still a poorly understood topic. In this paper, we analyze the \"gradient descent\" form of GAN optimization i.e., the natural setting where we simultaneously take small gradient steps in both generator and discriminator parameters. We show that even though GAN optimization does not correspond to a convex-concave game (even for simple parameterizations), under proper conditions, equilibrium points of this optimization procedure are still \\emph{locally asymptotically stable} for the traditional GAN f"},"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":"1706.04156","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-06-13T16:49:13Z","cross_cats_sorted":["cs.AI","math.OC","stat.ML"],"title_canon_sha256":"8c49236a9028e18012c2bf511466cb60bb9a3944c7f4b6a97d0def007897e868","abstract_canon_sha256":"93d0a01c518de6bda1bdc41063ff53bae337e01c1a4479ced7144f9000ca44a9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:26:08.004331Z","signature_b64":"jxVLKaocV1V80TlgiBoTtCBIShpJagYqVUSBK47lvzi6JBBmAIZc+ZnCGk2LtXa+3EZh36D4b7q9navz8b13CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dc4ba55ee818d692d5cfc68dae831ab9dd83a9d9bab67928739c85131eae7172","last_reissued_at":"2026-05-18T00:26:08.003668Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:26:08.003668Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Gradient descent GAN optimization is locally stable","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"J. Zico Kolter, Vaishnavh Nagarajan","submitted_at":"2017-06-13T16:49:13Z","abstract_excerpt":"Despite the growing prominence of generative adversarial networks (GANs), optimization in GANs is still a poorly understood topic. In this paper, we analyze the \"gradient descent\" form of GAN optimization i.e., the natural setting where we simultaneously take small gradient steps in both generator and discriminator parameters. We show that even though GAN optimization does not correspond to a convex-concave game (even for simple parameterizations), under proper conditions, equilibrium points of this optimization procedure are still \\emph{locally asymptotically stable} for the traditional GAN f"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1706.04156","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":"1706.04156","created_at":"2026-05-18T00:26:08.003763+00:00"},{"alias_kind":"arxiv_version","alias_value":"1706.04156v3","created_at":"2026-05-18T00:26:08.003763+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1706.04156","created_at":"2026-05-18T00:26:08.003763+00:00"},{"alias_kind":"pith_short_12","alias_value":"3RF2KXXIDDLJ","created_at":"2026-05-18T12:30:58.224056+00:00"},{"alias_kind":"pith_short_16","alias_value":"3RF2KXXIDDLJFVOP","created_at":"2026-05-18T12:30:58.224056+00:00"},{"alias_kind":"pith_short_8","alias_value":"3RF2KXXI","created_at":"2026-05-18T12:30:58.224056+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.11318","citing_title":"Nested Annealed Training Scheme for Generative Adversarial Networks","ref_index":52,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3RF2KXXIDDLJFVOPY2G25AY2XH","json":"https://pith.science/pith/3RF2KXXIDDLJFVOPY2G25AY2XH.json","graph_json":"https://pith.science/api/pith-number/3RF2KXXIDDLJFVOPY2G25AY2XH/graph.json","events_json":"https://pith.science/api/pith-number/3RF2KXXIDDLJFVOPY2G25AY2XH/events.json","paper":"https://pith.science/paper/3RF2KXXI"},"agent_actions":{"view_html":"https://pith.science/pith/3RF2KXXIDDLJFVOPY2G25AY2XH","download_json":"https://pith.science/pith/3RF2KXXIDDLJFVOPY2G25AY2XH.json","view_paper":"https://pith.science/paper/3RF2KXXI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1706.04156&json=true","fetch_graph":"https://pith.science/api/pith-number/3RF2KXXIDDLJFVOPY2G25AY2XH/graph.json","fetch_events":"https://pith.science/api/pith-number/3RF2KXXIDDLJFVOPY2G25AY2XH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3RF2KXXIDDLJFVOPY2G25AY2XH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3RF2KXXIDDLJFVOPY2G25AY2XH/action/storage_attestation","attest_author":"https://pith.science/pith/3RF2KXXIDDLJFVOPY2G25AY2XH/action/author_attestation","sign_citation":"https://pith.science/pith/3RF2KXXIDDLJFVOPY2G25AY2XH/action/citation_signature","submit_replication":"https://pith.science/pith/3RF2KXXIDDLJFVOPY2G25AY2XH/action/replication_record"}},"created_at":"2026-05-18T00:26:08.003763+00:00","updated_at":"2026-05-18T00:26:08.003763+00:00"}