{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:SUYCJ6GHG7NJBPVTIILGIA5WDR","short_pith_number":"pith:SUYCJ6GH","canonical_record":{"source":{"id":"2510.18183","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-10-21T00:14:45Z","cross_cats_sorted":["cs.GT"],"title_canon_sha256":"2cf02e02ed8fa40a8014519283e85d0ede184614cc5573f5f861fefa51474f5d","abstract_canon_sha256":"1b92b9c49787d3e25ad57d74b2cd13c64e17ec5b2c97973ab5861ec5917a0778"},"schema_version":"1.0"},"canonical_sha256":"953024f8c737da90beb342166403b61c698b4695b37a2e11ac81a134a0514fc5","source":{"kind":"arxiv","id":"2510.18183","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2510.18183","created_at":"2026-08-04T01:58:05Z"},{"alias_kind":"arxiv_version","alias_value":"2510.18183v3","created_at":"2026-08-04T01:58:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.18183","created_at":"2026-08-04T01:58:05Z"},{"alias_kind":"pith_short_12","alias_value":"SUYCJ6GHG7NJ","created_at":"2026-08-04T01:58:05Z"},{"alias_kind":"pith_short_16","alias_value":"SUYCJ6GHG7NJBPVT","created_at":"2026-08-04T01:58:05Z"},{"alias_kind":"pith_short_8","alias_value":"SUYCJ6GH","created_at":"2026-08-04T01:58:05Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:SUYCJ6GHG7NJBPVTIILGIA5WDR","target":"record","payload":{"canonical_record":{"source":{"id":"2510.18183","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-10-21T00:14:45Z","cross_cats_sorted":["cs.GT"],"title_canon_sha256":"2cf02e02ed8fa40a8014519283e85d0ede184614cc5573f5f861fefa51474f5d","abstract_canon_sha256":"1b92b9c49787d3e25ad57d74b2cd13c64e17ec5b2c97973ab5861ec5917a0778"},"schema_version":"1.0"},"canonical_sha256":"953024f8c737da90beb342166403b61c698b4695b37a2e11ac81a134a0514fc5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-04T01:58:05.660384Z","signature_b64":"BhDLzh7RQUhJ5lT/4fDRUsqKt6YzqINKbk9hN517mhuVxBUU17vWi6jOE+L6/Zi8+SaiJQjmgMd2X8jbWHobAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"953024f8c737da90beb342166403b61c698b4695b37a2e11ac81a134a0514fc5","last_reissued_at":"2026-08-04T01:58:05.658549Z","signature_status":"signed_v1","first_computed_at":"2026-08-04T01:58:05.658549Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2510.18183","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-04T01:58:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DawV2NmSz5YC4RoaoinsZ6ksTdDcqrW90Ihrxt4JIcAIjBAD1pYests8ssKGBZVLRsMpmRLwSDAr4MPS5SZqAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T01:46:27.428473Z"},"content_sha256":"e927bd2b95429aff51a87f7547d58062b97ac5fb73e572520a877912e8431cdd","schema_version":"1.0","event_id":"sha256:e927bd2b95429aff51a87f7547d58062b97ac5fb73e572520a877912e8431cdd"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:SUYCJ6GHG7NJBPVTIILGIA5WDR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"NashPG: A Policy Gradient Method with Iteratively Refined Regularization for Finding Nash Equilibria","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Placing iteratively refined regularization inside the policy-gradient objective guarantees strictly monotonic reduction in Bregman divergence toward Nash equilibria in two-player zero-sum extensive-form games.","cross_cats":["cs.GT"],"primary_cat":"cs.LG","authors_text":"Chang Xu, Cl\\'ement L. Canonne, Eason Yu, Nguyen H. Tran, Stefano V. Albrecht, Tzu Hao Liu, Yunke Wang","submitted_at":"2025-10-21T00:14:45Z","abstract_excerpt":"Finding Nash equilibria in two-player zero-sum imperfect-information games remains a central challenge in multi-agent reinforcement learning. Recent multi-round regularization methods offer a promising direction, yet existing approaches either require full enumeration of the game tree or rely on non-policy-gradient inner solvers that underperform in practice, leaving a scalable policy-gradient-based solution open. In this paper, we propose a novel multi-round regularization procedure and show that it guarantees strictly monotonic reduction in Bregman divergence to Nash equilibria and eventual "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"we propose a novel multi-round regularization procedure and show that it guarantees strictly monotonic reduction in Bregman divergence to Nash equilibria and eventual convergence to one in two-player zero-sum extensive-form games.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the regularization term can be placed directly inside a standard policy-gradient objective while preserving the strictly monotonic reduction property without requiring additional assumptions on the game tree or the inner solver.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"NashPG is a policy-gradient method with iteratively refined regularization that guarantees monotonic convergence to Nash equilibria in two-player zero-sum extensive-form games and scales to large benchmarks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Placing iteratively refined regularization inside the policy-gradient objective guarantees strictly monotonic reduction in Bregman divergence toward Nash equilibria in two-player zero-sum extensive-form games.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"d0965a035fd3bc3ea26ac7fd0495a1bcc4678833f6ca6dfa13aafd5f675aae7a"},"source":{"id":"2510.18183","kind":"arxiv","version":3},"verdict":{"id":"e47ffd0f-2cc7-4044-af0c-68dcac99cf8c","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-18T05:44:50.432118Z","strongest_claim":"we propose a novel multi-round regularization procedure and show that it guarantees strictly monotonic reduction in Bregman divergence to Nash equilibria and eventual convergence to one in two-player zero-sum extensive-form games.","one_line_summary":"NashPG is a policy-gradient method with iteratively refined regularization that guarantees monotonic convergence to Nash equilibria in two-player zero-sum extensive-form games and scales to large benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the regularization term can be placed directly inside a standard policy-gradient objective while preserving the strictly monotonic reduction property without requiring additional assumptions on the game tree or the inner solver.","pith_extraction_headline":"Placing iteratively refined regularization inside the policy-gradient objective guarantees strictly monotonic reduction in Bregman divergence toward Nash equilibria in two-player zero-sum extensive-form games."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2510.18183/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"},"verdict_id":"e47ffd0f-2cc7-4044-af0c-68dcac99cf8c"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-04T01:58:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"N1ym1UENJbiV9vmkrYMEOCSLrLxrSLuuINLkbWJU28vmMboZ+ozmn1sc6Nm5MwnVLTpg9fFzaXzrtpKQBmcgDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T01:46:27.429256Z"},"content_sha256":"c791737ae24b93e7444c4f5e956921d03a8acaa260012a123909a27e7924a1d9","schema_version":"1.0","event_id":"sha256:c791737ae24b93e7444c4f5e956921d03a8acaa260012a123909a27e7924a1d9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SUYCJ6GHG7NJBPVTIILGIA5WDR/bundle.json","state_url":"https://pith.science/pith/SUYCJ6GHG7NJBPVTIILGIA5WDR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SUYCJ6GHG7NJBPVTIILGIA5WDR/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":"2026-08-11T01:46:27Z","links":{"resolver":"https://pith.science/pith/SUYCJ6GHG7NJBPVTIILGIA5WDR","bundle":"https://pith.science/pith/SUYCJ6GHG7NJBPVTIILGIA5WDR/bundle.json","state":"https://pith.science/pith/SUYCJ6GHG7NJBPVTIILGIA5WDR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SUYCJ6GHG7NJBPVTIILGIA5WDR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:SUYCJ6GHG7NJBPVTIILGIA5WDR","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":"1b92b9c49787d3e25ad57d74b2cd13c64e17ec5b2c97973ab5861ec5917a0778","cross_cats_sorted":["cs.GT"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-10-21T00:14:45Z","title_canon_sha256":"2cf02e02ed8fa40a8014519283e85d0ede184614cc5573f5f861fefa51474f5d"},"schema_version":"1.0","source":{"id":"2510.18183","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2510.18183","created_at":"2026-08-04T01:58:05Z"},{"alias_kind":"arxiv_version","alias_value":"2510.18183v3","created_at":"2026-08-04T01:58:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.18183","created_at":"2026-08-04T01:58:05Z"},{"alias_kind":"pith_short_12","alias_value":"SUYCJ6GHG7NJ","created_at":"2026-08-04T01:58:05Z"},{"alias_kind":"pith_short_16","alias_value":"SUYCJ6GHG7NJBPVT","created_at":"2026-08-04T01:58:05Z"},{"alias_kind":"pith_short_8","alias_value":"SUYCJ6GH","created_at":"2026-08-04T01:58:05Z"}],"graph_snapshots":[{"event_id":"sha256:c791737ae24b93e7444c4f5e956921d03a8acaa260012a123909a27e7924a1d9","target":"graph","created_at":"2026-08-04T01:58:05Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"we propose a novel multi-round regularization procedure and show that it guarantees strictly monotonic reduction in Bregman divergence to Nash equilibria and eventual convergence to one in two-player zero-sum extensive-form games."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the regularization term can be placed directly inside a standard policy-gradient objective while preserving the strictly monotonic reduction property without requiring additional assumptions on the game tree or the inner solver."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"NashPG is a policy-gradient method with iteratively refined regularization that guarantees monotonic convergence to Nash equilibria in two-player zero-sum extensive-form games and scales to large benchmarks."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Placing iteratively refined regularization inside the policy-gradient objective guarantees strictly monotonic reduction in Bregman divergence toward Nash equilibria in two-player zero-sum extensive-form games."}],"snapshot_sha256":"d0965a035fd3bc3ea26ac7fd0495a1bcc4678833f6ca6dfa13aafd5f675aae7a"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2510.18183/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Finding Nash equilibria in two-player zero-sum imperfect-information games remains a central challenge in multi-agent reinforcement learning. Recent multi-round regularization methods offer a promising direction, yet existing approaches either require full enumeration of the game tree or rely on non-policy-gradient inner solvers that underperform in practice, leaving a scalable policy-gradient-based solution open. In this paper, we propose a novel multi-round regularization procedure and show that it guarantees strictly monotonic reduction in Bregman divergence to Nash equilibria and eventual ","authors_text":"Chang Xu, Cl\\'ement L. Canonne, Eason Yu, Nguyen H. Tran, Stefano V. Albrecht, Tzu Hao Liu, Yunke Wang","cross_cats":["cs.GT"],"headline":"Placing iteratively refined regularization inside the policy-gradient objective guarantees strictly monotonic reduction in Bregman divergence toward Nash equilibria in two-player zero-sum extensive-form games.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-10-21T00:14:45Z","title":"NashPG: A Policy Gradient Method with Iteratively Refined Regularization for Finding Nash Equilibria"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.18183","kind":"arxiv","version":3},"verdict":{"created_at":"2026-05-18T05:44:50.432118Z","id":"e47ffd0f-2cc7-4044-af0c-68dcac99cf8c","model_set":{"reader":"grok-4.3"},"one_line_summary":"NashPG is a policy-gradient method with iteratively refined regularization that guarantees monotonic convergence to Nash equilibria in two-player zero-sum extensive-form games and scales to large benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Placing iteratively refined regularization inside the policy-gradient objective guarantees strictly monotonic reduction in Bregman divergence toward Nash equilibria in two-player zero-sum extensive-form games.","strongest_claim":"we propose a novel multi-round regularization procedure and show that it guarantees strictly monotonic reduction in Bregman divergence to Nash equilibria and eventual convergence to one in two-player zero-sum extensive-form games.","weakest_assumption":"That the regularization term can be placed directly inside a standard policy-gradient objective while preserving the strictly monotonic reduction property without requiring additional assumptions on the game tree or the inner solver."}},"verdict_id":"e47ffd0f-2cc7-4044-af0c-68dcac99cf8c"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:e927bd2b95429aff51a87f7547d58062b97ac5fb73e572520a877912e8431cdd","target":"record","created_at":"2026-08-04T01:58:05Z","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":"1b92b9c49787d3e25ad57d74b2cd13c64e17ec5b2c97973ab5861ec5917a0778","cross_cats_sorted":["cs.GT"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-10-21T00:14:45Z","title_canon_sha256":"2cf02e02ed8fa40a8014519283e85d0ede184614cc5573f5f861fefa51474f5d"},"schema_version":"1.0","source":{"id":"2510.18183","kind":"arxiv","version":3}},"canonical_sha256":"953024f8c737da90beb342166403b61c698b4695b37a2e11ac81a134a0514fc5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"953024f8c737da90beb342166403b61c698b4695b37a2e11ac81a134a0514fc5","first_computed_at":"2026-08-04T01:58:05.658549Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-04T01:58:05.658549Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"BhDLzh7RQUhJ5lT/4fDRUsqKt6YzqINKbk9hN517mhuVxBUU17vWi6jOE+L6/Zi8+SaiJQjmgMd2X8jbWHobAA==","signature_status":"signed_v1","signed_at":"2026-08-04T01:58:05.660384Z","signed_message":"canonical_sha256_bytes"},"source_id":"2510.18183","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e927bd2b95429aff51a87f7547d58062b97ac5fb73e572520a877912e8431cdd","sha256:c791737ae24b93e7444c4f5e956921d03a8acaa260012a123909a27e7924a1d9"],"state_sha256":"a0d44d3d301ee7744f984fd26b9aa142a71ed6608c8ce888040ff1e57ac8ef95"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YcK992ZptNd6kfSJNdApxKDDN+3/4gXCgYnxeV9IRNaq9ZGmUbG3qSYkKD+YEVV4/2tN3psbYxq/0T56z2cYDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T01:46:27.433349Z","bundle_sha256":"3b6e71ec9a7b48d81d54d397aae534e555a6d40216adf40d2146e8607959e31d"}}