{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:TSMERPH4KLGHNB7TH4334KW2C6","short_pith_number":"pith:TSMERPH4","canonical_record":{"source":{"id":"1806.04498","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2018-06-12T13:27:23Z","cross_cats_sorted":["cs.CV","cs.LG"],"title_canon_sha256":"49412633c6279f1fa28e9c093d41ecefa422c482e69f55da6116325a174b4b04","abstract_canon_sha256":"5dcbdece5e418d5fe2f44995cd05322d04eb462d276ca950ffdfe55b4ab2dc6f"},"schema_version":"1.0"},"canonical_sha256":"9c9848bcfc52cc7687f33f37be2ada178e304422fe4f60cac0a62a1f8c05354f","source":{"kind":"arxiv","id":"1806.04498","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1806.04498","created_at":"2026-05-17T23:52:41Z"},{"alias_kind":"arxiv_version","alias_value":"1806.04498v2","created_at":"2026-05-17T23:52:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1806.04498","created_at":"2026-05-17T23:52:41Z"},{"alias_kind":"pith_short_12","alias_value":"TSMERPH4KLGH","created_at":"2026-05-18T12:32:56Z"},{"alias_kind":"pith_short_16","alias_value":"TSMERPH4KLGHNB7T","created_at":"2026-05-18T12:32:56Z"},{"alias_kind":"pith_short_8","alias_value":"TSMERPH4","created_at":"2026-05-18T12:32:56Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:TSMERPH4KLGHNB7TH4334KW2C6","target":"record","payload":{"canonical_record":{"source":{"id":"1806.04498","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2018-06-12T13:27:23Z","cross_cats_sorted":["cs.CV","cs.LG"],"title_canon_sha256":"49412633c6279f1fa28e9c093d41ecefa422c482e69f55da6116325a174b4b04","abstract_canon_sha256":"5dcbdece5e418d5fe2f44995cd05322d04eb462d276ca950ffdfe55b4ab2dc6f"},"schema_version":"1.0"},"canonical_sha256":"9c9848bcfc52cc7687f33f37be2ada178e304422fe4f60cac0a62a1f8c05354f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:52:41.792875Z","signature_b64":"SdmoH9aq7gbgKjYw8YO1rZTMaQVFtxgjQ6/9Sla0NqCCMFd3fuIMPJ3E9z0C8kaSAKcLFSkANJ+J6n7DAA1vBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9c9848bcfc52cc7687f33f37be2ada178e304422fe4f60cac0a62a1f8c05354f","last_reissued_at":"2026-05-17T23:52:41.792213Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:52:41.792213Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1806.04498","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:52:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"86S7HHouem3tKBQ7e68emq38kM7VRCTGiUWoa0oOsIqYLBnqmnOjeJ/YsJuiGqKGGnZmiaXvJ4ZuDp6EaEYUDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-02T19:28:36.555417Z"},"content_sha256":"50adffe4b6eafb532cee9aecccfa6a15368baa32217e64b3b29b2558b3aa5dbf","schema_version":"1.0","event_id":"sha256:50adffe4b6eafb532cee9aecccfa6a15368baa32217e64b3b29b2558b3aa5dbf"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:TSMERPH4KLGHNB7TH4334KW2C6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"The Unusual Effectiveness of Averaging in GAN Training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","cs.LG"],"primary_cat":"stat.ML","authors_text":"Chuan-Sheng Foo, Georgios Piliouras, Kim-Hui Yap, Stefan Winkler, Vijay Chandrasekhar, Yasin Yaz{\\i}c{\\i}","submitted_at":"2018-06-12T13:27:23Z","abstract_excerpt":"We examine two different techniques for parameter averaging in GAN training. Moving Average (MA) computes the time-average of parameters, whereas Exponential Moving Average (EMA) computes an exponentially discounted sum. Whilst MA is known to lead to convergence in bilinear settings, we provide the -- to our knowledge -- first theoretical arguments in support of EMA. We show that EMA converges to limit cycles around the equilibrium with vanishing amplitude as the discount parameter approaches one for simple bilinear games and also enhances the stability of general GAN training. We establish ex"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1806.04498","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":""},"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:52:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HszaUo7BihDM6Sbs1WqE/o270MhLMkSglHISTqcrtXEbwPOs5UQ847ABFBL0B0vlFl+/hB6hAR4J4p5mKLU9Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-02T19:28:36.555771Z"},"content_sha256":"628ac5a2553b760be46527778cd74bf524606c5d792f1bafad5f442b2d197665","schema_version":"1.0","event_id":"sha256:628ac5a2553b760be46527778cd74bf524606c5d792f1bafad5f442b2d197665"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TSMERPH4KLGHNB7TH4334KW2C6/bundle.json","state_url":"https://pith.science/pith/TSMERPH4KLGHNB7TH4334KW2C6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TSMERPH4KLGHNB7TH4334KW2C6/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-02T19:28:36Z","links":{"resolver":"https://pith.science/pith/TSMERPH4KLGHNB7TH4334KW2C6","bundle":"https://pith.science/pith/TSMERPH4KLGHNB7TH4334KW2C6/bundle.json","state":"https://pith.science/pith/TSMERPH4KLGHNB7TH4334KW2C6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TSMERPH4KLGHNB7TH4334KW2C6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:TSMERPH4KLGHNB7TH4334KW2C6","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"5dcbdece5e418d5fe2f44995cd05322d04eb462d276ca950ffdfe55b4ab2dc6f","cross_cats_sorted":["cs.CV","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2018-06-12T13:27:23Z","title_canon_sha256":"49412633c6279f1fa28e9c093d41ecefa422c482e69f55da6116325a174b4b04"},"schema_version":"1.0","source":{"id":"1806.04498","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1806.04498","created_at":"2026-05-17T23:52:41Z"},{"alias_kind":"arxiv_version","alias_value":"1806.04498v2","created_at":"2026-05-17T23:52:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1806.04498","created_at":"2026-05-17T23:52:41Z"},{"alias_kind":"pith_short_12","alias_value":"TSMERPH4KLGH","created_at":"2026-05-18T12:32:56Z"},{"alias_kind":"pith_short_16","alias_value":"TSMERPH4KLGHNB7T","created_at":"2026-05-18T12:32:56Z"},{"alias_kind":"pith_short_8","alias_value":"TSMERPH4","created_at":"2026-05-18T12:32:56Z"}],"graph_snapshots":[{"event_id":"sha256:628ac5a2553b760be46527778cd74bf524606c5d792f1bafad5f442b2d197665","target":"graph","created_at":"2026-05-17T23:52:41Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"We examine two different techniques for parameter averaging in GAN training. Moving Average (MA) computes the time-average of parameters, whereas Exponential Moving Average (EMA) computes an exponentially discounted sum. Whilst MA is known to lead to convergence in bilinear settings, we provide the -- to our knowledge -- first theoretical arguments in support of EMA. We show that EMA converges to limit cycles around the equilibrium with vanishing amplitude as the discount parameter approaches one for simple bilinear games and also enhances the stability of general GAN training. We establish ex","authors_text":"Chuan-Sheng Foo, Georgios Piliouras, Kim-Hui Yap, Stefan Winkler, Vijay Chandrasekhar, Yasin Yaz{\\i}c{\\i}","cross_cats":["cs.CV","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2018-06-12T13:27:23Z","title":"The Unusual Effectiveness of Averaging in GAN Training"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1806.04498","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:50adffe4b6eafb532cee9aecccfa6a15368baa32217e64b3b29b2558b3aa5dbf","target":"record","created_at":"2026-05-17T23:52:41Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"5dcbdece5e418d5fe2f44995cd05322d04eb462d276ca950ffdfe55b4ab2dc6f","cross_cats_sorted":["cs.CV","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2018-06-12T13:27:23Z","title_canon_sha256":"49412633c6279f1fa28e9c093d41ecefa422c482e69f55da6116325a174b4b04"},"schema_version":"1.0","source":{"id":"1806.04498","kind":"arxiv","version":2}},"canonical_sha256":"9c9848bcfc52cc7687f33f37be2ada178e304422fe4f60cac0a62a1f8c05354f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9c9848bcfc52cc7687f33f37be2ada178e304422fe4f60cac0a62a1f8c05354f","first_computed_at":"2026-05-17T23:52:41.792213Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:52:41.792213Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"SdmoH9aq7gbgKjYw8YO1rZTMaQVFtxgjQ6/9Sla0NqCCMFd3fuIMPJ3E9z0C8kaSAKcLFSkANJ+J6n7DAA1vBA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:52:41.792875Z","signed_message":"canonical_sha256_bytes"},"source_id":"1806.04498","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:50adffe4b6eafb532cee9aecccfa6a15368baa32217e64b3b29b2558b3aa5dbf","sha256:628ac5a2553b760be46527778cd74bf524606c5d792f1bafad5f442b2d197665"],"state_sha256":"32f94082ed937e51b5cf50b9fcc6bc112dc463d7e9221b60d879aaa3024fcbaf"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1c2DCvGeaVRdCgsAwdpOhb6rAvezJFqiouIAtmhospJ3xgbXnRioD5+kaJ056Yg1yA71yndZgj67rS2IgwOLAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-02T19:28:36.557791Z","bundle_sha256":"cc34ec529750240c1fa146d4c93740c28dadebd54cfa93c510984082f83d069e"}}