{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:BCNWL6PEOTYXEV3T4VTQ4TTZUL","short_pith_number":"pith:BCNWL6PE","schema_version":"1.0","canonical_sha256":"089b65f9e474f1725773e5670e4e79a2eab3c09245fb8bd2d5f957db0d4bb7bd","source":{"kind":"arxiv","id":"2008.01825","version":2},"attestation_state":"computed","paper":{"title":"Robust Reinforcement Learning using Adversarial Populations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.MA","cs.RO","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexandre Bayen, Eugene Vinitsky, Kanaad Parvate, Kathy Jang, Pieter Abbeel, Yuqing Du","submitted_at":"2020-08-04T20:57:32Z","abstract_excerpt":"Reinforcement Learning (RL) is an effective tool for controller design but can struggle with issues of robustness, failing catastrophically when the underlying system dynamics are perturbed. The Robust RL formulation tackles this by adding worst-case adversarial noise to the dynamics and constructing the noise distribution as the solution to a zero-sum minimax game. However, existing work on learning solutions to the Robust RL formulation has primarily focused on training a single RL agent against a single adversary. In this work, we demonstrate that using a single adversary does not consisten"},"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":"2008.01825","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-08-04T20:57:32Z","cross_cats_sorted":["cs.MA","cs.RO","stat.ML"],"title_canon_sha256":"f1536e511b0f18577d9967b38277e682400fc4e3182a5213826425b7874daae5","abstract_canon_sha256":"7283bb19c1da65847229dedad8ddf0af26dd32f38cf9b8630499ff83135f2320"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:37:24.824045Z","signature_b64":"wcSFwohzktQSpQGXt1X8rOFaRpzaZBQymb8Plb6xuyFnnYAn34gWazUNnJpHoc7yt5v4cwjshB/imG+wCsI8Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"089b65f9e474f1725773e5670e4e79a2eab3c09245fb8bd2d5f957db0d4bb7bd","last_reissued_at":"2026-07-05T01:37:24.823596Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:37:24.823596Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Robust Reinforcement Learning using Adversarial Populations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.MA","cs.RO","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexandre Bayen, Eugene Vinitsky, Kanaad Parvate, Kathy Jang, Pieter Abbeel, Yuqing Du","submitted_at":"2020-08-04T20:57:32Z","abstract_excerpt":"Reinforcement Learning (RL) is an effective tool for controller design but can struggle with issues of robustness, failing catastrophically when the underlying system dynamics are perturbed. The Robust RL formulation tackles this by adding worst-case adversarial noise to the dynamics and constructing the noise distribution as the solution to a zero-sum minimax game. However, existing work on learning solutions to the Robust RL formulation has primarily focused on training a single RL agent against a single adversary. In this work, we demonstrate that using a single adversary does not consisten"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2008.01825","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/2008.01825/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":"2008.01825","created_at":"2026-07-05T01:37:24.823654+00:00"},{"alias_kind":"arxiv_version","alias_value":"2008.01825v2","created_at":"2026-07-05T01:37:24.823654+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2008.01825","created_at":"2026-07-05T01:37:24.823654+00:00"},{"alias_kind":"pith_short_12","alias_value":"BCNWL6PEOTYX","created_at":"2026-07-05T01:37:24.823654+00:00"},{"alias_kind":"pith_short_16","alias_value":"BCNWL6PEOTYXEV3T","created_at":"2026-07-05T01:37:24.823654+00:00"},{"alias_kind":"pith_short_8","alias_value":"BCNWL6PE","created_at":"2026-07-05T01:37:24.823654+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22579","citing_title":"Stationary Robust Mean-Field Games under Model Mismatches","ref_index":65,"is_internal_anchor":false},{"citing_arxiv_id":"2502.02844","citing_title":"Wolfpack Adversarial Attack for Robust Multi-Agent Reinforcement Learning","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2602.08813","citing_title":"Robust Policy Optimization to Prevent Catastrophic Forgetting","ref_index":53,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10974","citing_title":"Robust Adversarial Policy Optimization Under Dynamics Uncertainty","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14243","citing_title":"Optimistic Policy Learning under Pessimistic Adversaries with Regret and Violation Guarantees","ref_index":20,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BCNWL6PEOTYXEV3T4VTQ4TTZUL","json":"https://pith.science/pith/BCNWL6PEOTYXEV3T4VTQ4TTZUL.json","graph_json":"https://pith.science/api/pith-number/BCNWL6PEOTYXEV3T4VTQ4TTZUL/graph.json","events_json":"https://pith.science/api/pith-number/BCNWL6PEOTYXEV3T4VTQ4TTZUL/events.json","paper":"https://pith.science/paper/BCNWL6PE"},"agent_actions":{"view_html":"https://pith.science/pith/BCNWL6PEOTYXEV3T4VTQ4TTZUL","download_json":"https://pith.science/pith/BCNWL6PEOTYXEV3T4VTQ4TTZUL.json","view_paper":"https://pith.science/paper/BCNWL6PE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2008.01825&json=true","fetch_graph":"https://pith.science/api/pith-number/BCNWL6PEOTYXEV3T4VTQ4TTZUL/graph.json","fetch_events":"https://pith.science/api/pith-number/BCNWL6PEOTYXEV3T4VTQ4TTZUL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BCNWL6PEOTYXEV3T4VTQ4TTZUL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BCNWL6PEOTYXEV3T4VTQ4TTZUL/action/storage_attestation","attest_author":"https://pith.science/pith/BCNWL6PEOTYXEV3T4VTQ4TTZUL/action/author_attestation","sign_citation":"https://pith.science/pith/BCNWL6PEOTYXEV3T4VTQ4TTZUL/action/citation_signature","submit_replication":"https://pith.science/pith/BCNWL6PEOTYXEV3T4VTQ4TTZUL/action/replication_record"}},"created_at":"2026-07-05T01:37:24.823654+00:00","updated_at":"2026-07-05T01:37:24.823654+00:00"}