{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:4B55JHO33KPQWNYRH7GAL42HRH","short_pith_number":"pith:4B55JHO3","schema_version":"1.0","canonical_sha256":"e07bd49ddbda9f0b37113fcc05f34789df4fd80f2f7cb1d8e26daa18c9145159","source":{"kind":"arxiv","id":"1912.00953","version":2},"attestation_state":"computed","paper":{"title":"LOGAN: Latent Optimisation for Generative Adversarial Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"David Balduzzi, Jeff Donahue, Karen Simonyan, Timothy Lillicrap, Yan Wu","submitted_at":"2019-12-02T17:30:05Z","abstract_excerpt":"Training generative adversarial networks requires balancing of delicate adversarial dynamics. Even with careful tuning, training may diverge or end up in a bad equilibrium with dropped modes. In this work, we improve CS-GAN with natural gradient-based latent optimisation and show that it improves adversarial dynamics by enhancing interactions between the discriminator and the generator. Our experiments demonstrate that latent optimisation can significantly improve GAN training, obtaining state-of-the-art performance for the ImageNet ($128 \\times 128$) dataset. Our model achieves an Inception S"},"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":"1912.00953","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-12-02T17:30:05Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"5ce90b21743be810c3854c58a40193caa6c3cd174d01da2f81fa330703889ca4","abstract_canon_sha256":"e991af3312f0f743fc9740d6d084fca069566c6234163104b1556b234dab3122"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:15:23.592523Z","signature_b64":"vBLO4uW1a7uhwQNq0tM8jLvi2Q1/3XVPBQpF9G1KOqx77f+xUg/f3/WuUVARsKoTUUcUFkyRUXV7Sv9VjIadCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e07bd49ddbda9f0b37113fcc05f34789df4fd80f2f7cb1d8e26daa18c9145159","last_reissued_at":"2026-07-05T01:15:23.591957Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:15:23.591957Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LOGAN: Latent Optimisation for Generative Adversarial Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"David Balduzzi, Jeff Donahue, Karen Simonyan, Timothy Lillicrap, Yan Wu","submitted_at":"2019-12-02T17:30:05Z","abstract_excerpt":"Training generative adversarial networks requires balancing of delicate adversarial dynamics. Even with careful tuning, training may diverge or end up in a bad equilibrium with dropped modes. In this work, we improve CS-GAN with natural gradient-based latent optimisation and show that it improves adversarial dynamics by enhancing interactions between the discriminator and the generator. Our experiments demonstrate that latent optimisation can significantly improve GAN training, obtaining state-of-the-art performance for the ImageNet ($128 \\times 128$) dataset. Our model achieves an Inception S"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1912.00953","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/1912.00953/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":"1912.00953","created_at":"2026-07-05T01:15:23.592012+00:00"},{"alias_kind":"arxiv_version","alias_value":"1912.00953v2","created_at":"2026-07-05T01:15:23.592012+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1912.00953","created_at":"2026-07-05T01:15:23.592012+00:00"},{"alias_kind":"pith_short_12","alias_value":"4B55JHO33KPQ","created_at":"2026-07-05T01:15:23.592012+00:00"},{"alias_kind":"pith_short_16","alias_value":"4B55JHO33KPQWNYR","created_at":"2026-07-05T01:15:23.592012+00:00"},{"alias_kind":"pith_short_8","alias_value":"4B55JHO3","created_at":"2026-07-05T01:15:23.592012+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2602.12280","citing_title":"Stroke of Surprise: Progressive Semantic Illusions in Vector Sketching","ref_index":123,"is_internal_anchor":false},{"citing_arxiv_id":"2105.05233","citing_title":"Diffusion Models Beat GANs on Image Synthesis","ref_index":68,"is_internal_anchor":false},{"citing_arxiv_id":"2207.12598","citing_title":"Classifier-Free Diffusion Guidance","ref_index":23,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4B55JHO33KPQWNYRH7GAL42HRH","json":"https://pith.science/pith/4B55JHO33KPQWNYRH7GAL42HRH.json","graph_json":"https://pith.science/api/pith-number/4B55JHO33KPQWNYRH7GAL42HRH/graph.json","events_json":"https://pith.science/api/pith-number/4B55JHO33KPQWNYRH7GAL42HRH/events.json","paper":"https://pith.science/paper/4B55JHO3"},"agent_actions":{"view_html":"https://pith.science/pith/4B55JHO33KPQWNYRH7GAL42HRH","download_json":"https://pith.science/pith/4B55JHO33KPQWNYRH7GAL42HRH.json","view_paper":"https://pith.science/paper/4B55JHO3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1912.00953&json=true","fetch_graph":"https://pith.science/api/pith-number/4B55JHO33KPQWNYRH7GAL42HRH/graph.json","fetch_events":"https://pith.science/api/pith-number/4B55JHO33KPQWNYRH7GAL42HRH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4B55JHO33KPQWNYRH7GAL42HRH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4B55JHO33KPQWNYRH7GAL42HRH/action/storage_attestation","attest_author":"https://pith.science/pith/4B55JHO33KPQWNYRH7GAL42HRH/action/author_attestation","sign_citation":"https://pith.science/pith/4B55JHO33KPQWNYRH7GAL42HRH/action/citation_signature","submit_replication":"https://pith.science/pith/4B55JHO33KPQWNYRH7GAL42HRH/action/replication_record"}},"created_at":"2026-07-05T01:15:23.592012+00:00","updated_at":"2026-07-05T01:15:23.592012+00:00"}