{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:64LUOPCPKA2M3CSJW47FPIOOYR","short_pith_number":"pith:64LUOPCP","schema_version":"1.0","canonical_sha256":"f717473c4f5034cd8a49b73e57a1cec4403d7168c9c79a52e0f84056bb4a972c","source":{"kind":"arxiv","id":"1704.00028","version":3},"attestation_state":"computed","paper":{"title":"Improved Training of Wasserstein GANs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Aaron Courville, Faruk Ahmed, Ishaan Gulrajani, Martin Arjovsky, Vincent Dumoulin","submitted_at":"2017-03-31T19:25:00Z","abstract_excerpt":"Generative Adversarial Networks (GANs) are powerful generative models, but suffer from training instability. The recently proposed Wasserstein GAN (WGAN) makes progress toward stable training of GANs, but sometimes can still generate only low-quality samples or fail to converge. We find that these problems are often due to the use of weight clipping in WGAN to enforce a Lipschitz constraint on the critic, which can lead to undesired behavior. We propose an alternative to clipping weights: penalize the norm of gradient of the critic with respect to its input. Our proposed method performs better"},"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":"1704.00028","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-03-31T19:25:00Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"21ef08626ae5c051b25ac81e07ecea75bdee32fb343913a18e81a67289bb216e","abstract_canon_sha256":"e54d48cac72771a693764a435117a2d30c27b1a5bf22144612d8cb9bcd46bc74"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:27:21.598927Z","signature_b64":"ptsHnQM7+sQlMuuZoNcRMguikoPmSLZSMARsB99E1+f0LY/bjBNUT3xMyQtS6oA4sro0tbauQ6MwVrhOVzbJAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f717473c4f5034cd8a49b73e57a1cec4403d7168c9c79a52e0f84056bb4a972c","last_reissued_at":"2026-05-18T00:27:21.598303Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:27:21.598303Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improved Training of Wasserstein GANs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Aaron Courville, Faruk Ahmed, Ishaan Gulrajani, Martin Arjovsky, Vincent Dumoulin","submitted_at":"2017-03-31T19:25:00Z","abstract_excerpt":"Generative Adversarial Networks (GANs) are powerful generative models, but suffer from training instability. The recently proposed Wasserstein GAN (WGAN) makes progress toward stable training of GANs, but sometimes can still generate only low-quality samples or fail to converge. We find that these problems are often due to the use of weight clipping in WGAN to enforce a Lipschitz constraint on the critic, which can lead to undesired behavior. We propose an alternative to clipping weights: penalize the norm of gradient of the critic with respect to its input. Our proposed method performs better"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1704.00028","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":"1704.00028","created_at":"2026-05-18T00:27:21.598400+00:00"},{"alias_kind":"arxiv_version","alias_value":"1704.00028v3","created_at":"2026-05-18T00:27:21.598400+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1704.00028","created_at":"2026-05-18T00:27:21.598400+00:00"},{"alias_kind":"pith_short_12","alias_value":"64LUOPCPKA2M","created_at":"2026-05-18T12:31:03.183658+00:00"},{"alias_kind":"pith_short_16","alias_value":"64LUOPCPKA2M3CSJ","created_at":"2026-05-18T12:31:03.183658+00:00"},{"alias_kind":"pith_short_8","alias_value":"64LUOPCP","created_at":"2026-05-18T12:31:03.183658+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":12,"internal_anchor_count":7,"sample":[{"citing_arxiv_id":"1907.03355","citing_title":"Improving Detection of Credit Card Fraudulent Transactions using Generative Adversarial Networks","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18404","citing_title":"JanusPipe: Efficient Pipeline Parallel Training for Machine Learning Interatomic Potentials","ref_index":43,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18404","citing_title":"JanusPipe: Efficient Pipeline Parallel Training for Machine Learning Interatomic Potentials","ref_index":43,"is_internal_anchor":true},{"citing_arxiv_id":"2508.07798","citing_title":"Generative Inversion for Property-Targeted Materials Design: Application to Shape Memory Alloys","ref_index":53,"is_internal_anchor":true},{"citing_arxiv_id":"2509.00931","citing_title":"Semi-Supervised Bayesian GANs with Log-Signatures for Uncertainty-Aware Credit Card Fraud Detection","ref_index":43,"is_internal_anchor":true},{"citing_arxiv_id":"2603.21599","citing_title":"Conditional Wasserstein GAN for Simulating Neutrino Event Summaries using Incident Energy of Electron Neutrinos","ref_index":42,"is_internal_anchor":true},{"citing_arxiv_id":"1801.01401","citing_title":"Demystifying MMD GANs","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"1710.10196","citing_title":"Progressive Growing of GANs for Improved Quality, Stability, and Variation","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09004","citing_title":"Separate Universe Super-Resolution Emulator","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07860","citing_title":"On the Tradeoffs of On-Device Generative Models in Federated Predictive Maintenance Systems","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07565","citing_title":"Ensemble Distributionally Robust Bayesian Optimisation","ref_index":112,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13191","citing_title":"Fast Voxelization and Level of Detail for Microgeometry Rendering","ref_index":56,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/64LUOPCPKA2M3CSJW47FPIOOYR","json":"https://pith.science/pith/64LUOPCPKA2M3CSJW47FPIOOYR.json","graph_json":"https://pith.science/api/pith-number/64LUOPCPKA2M3CSJW47FPIOOYR/graph.json","events_json":"https://pith.science/api/pith-number/64LUOPCPKA2M3CSJW47FPIOOYR/events.json","paper":"https://pith.science/paper/64LUOPCP"},"agent_actions":{"view_html":"https://pith.science/pith/64LUOPCPKA2M3CSJW47FPIOOYR","download_json":"https://pith.science/pith/64LUOPCPKA2M3CSJW47FPIOOYR.json","view_paper":"https://pith.science/paper/64LUOPCP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1704.00028&json=true","fetch_graph":"https://pith.science/api/pith-number/64LUOPCPKA2M3CSJW47FPIOOYR/graph.json","fetch_events":"https://pith.science/api/pith-number/64LUOPCPKA2M3CSJW47FPIOOYR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/64LUOPCPKA2M3CSJW47FPIOOYR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/64LUOPCPKA2M3CSJW47FPIOOYR/action/storage_attestation","attest_author":"https://pith.science/pith/64LUOPCPKA2M3CSJW47FPIOOYR/action/author_attestation","sign_citation":"https://pith.science/pith/64LUOPCPKA2M3CSJW47FPIOOYR/action/citation_signature","submit_replication":"https://pith.science/pith/64LUOPCPKA2M3CSJW47FPIOOYR/action/replication_record"}},"created_at":"2026-05-18T00:27:21.598400+00:00","updated_at":"2026-05-18T00:27:21.598400+00:00"}