{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:3FS5KGFDLWMT4ZNZRLFM3CCO4I","short_pith_number":"pith:3FS5KGFD","schema_version":"1.0","canonical_sha256":"d965d518a35d993e65b98acacd884ee22410b66cc2a9b0dbe82844fefd9c8cc9","source":{"kind":"arxiv","id":"2102.12769","version":1},"attestation_state":"computed","paper":{"title":"No-Regret Reinforcement Learning with Heavy-Tailed Rewards","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Vincent Zhuang, Yanan Sui","submitted_at":"2021-02-25T10:25:57Z","abstract_excerpt":"Reinforcement learning algorithms typically assume rewards to be sampled from light-tailed distributions, such as Gaussian or bounded. However, a wide variety of real-world systems generate rewards that follow heavy-tailed distributions. We consider such scenarios in the setting of undiscounted reinforcement learning. By constructing a lower bound, we show that the difficulty of learning heavy-tailed rewards asymptotically dominates the difficulty of learning transition probabilities. Leveraging techniques from robust mean estimation, we propose Heavy-UCRL2 and Heavy-Q-Learning, and show that "},"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":"2102.12769","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-02-25T10:25:57Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"d584e0a4a2831408d614bcfa6d85c9eabd9833020e9398dc75b4d64a282e4e5c","abstract_canon_sha256":"99eb1f0090ef804fef53dacaf17251258fe0d91f5aaaa46e47784e331e168b19"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:18:23.634769Z","signature_b64":"dkj//KQgvZK7oA1yhHs9QM5CdH0Mu1CnRAaezd+y+7eziV0UqE91fCwDOoDF9uk0P8ZCF7VbFf5bgKl7IXMeAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d965d518a35d993e65b98acacd884ee22410b66cc2a9b0dbe82844fefd9c8cc9","last_reissued_at":"2026-07-05T02:18:23.634312Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:18:23.634312Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"No-Regret Reinforcement Learning with Heavy-Tailed Rewards","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Vincent Zhuang, Yanan Sui","submitted_at":"2021-02-25T10:25:57Z","abstract_excerpt":"Reinforcement learning algorithms typically assume rewards to be sampled from light-tailed distributions, such as Gaussian or bounded. However, a wide variety of real-world systems generate rewards that follow heavy-tailed distributions. We consider such scenarios in the setting of undiscounted reinforcement learning. By constructing a lower bound, we show that the difficulty of learning heavy-tailed rewards asymptotically dominates the difficulty of learning transition probabilities. Leveraging techniques from robust mean estimation, we propose Heavy-UCRL2 and Heavy-Q-Learning, and show that "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2102.12769","kind":"arxiv","version":1},"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/2102.12769/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":"2102.12769","created_at":"2026-07-05T02:18:23.634388+00:00"},{"alias_kind":"arxiv_version","alias_value":"2102.12769v1","created_at":"2026-07-05T02:18:23.634388+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2102.12769","created_at":"2026-07-05T02:18:23.634388+00:00"},{"alias_kind":"pith_short_12","alias_value":"3FS5KGFDLWMT","created_at":"2026-07-05T02:18:23.634388+00:00"},{"alias_kind":"pith_short_16","alias_value":"3FS5KGFDLWMT4ZNZ","created_at":"2026-07-05T02:18:23.634388+00:00"},{"alias_kind":"pith_short_8","alias_value":"3FS5KGFD","created_at":"2026-07-05T02:18:23.634388+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3FS5KGFDLWMT4ZNZRLFM3CCO4I","json":"https://pith.science/pith/3FS5KGFDLWMT4ZNZRLFM3CCO4I.json","graph_json":"https://pith.science/api/pith-number/3FS5KGFDLWMT4ZNZRLFM3CCO4I/graph.json","events_json":"https://pith.science/api/pith-number/3FS5KGFDLWMT4ZNZRLFM3CCO4I/events.json","paper":"https://pith.science/paper/3FS5KGFD"},"agent_actions":{"view_html":"https://pith.science/pith/3FS5KGFDLWMT4ZNZRLFM3CCO4I","download_json":"https://pith.science/pith/3FS5KGFDLWMT4ZNZRLFM3CCO4I.json","view_paper":"https://pith.science/paper/3FS5KGFD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2102.12769&json=true","fetch_graph":"https://pith.science/api/pith-number/3FS5KGFDLWMT4ZNZRLFM3CCO4I/graph.json","fetch_events":"https://pith.science/api/pith-number/3FS5KGFDLWMT4ZNZRLFM3CCO4I/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3FS5KGFDLWMT4ZNZRLFM3CCO4I/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3FS5KGFDLWMT4ZNZRLFM3CCO4I/action/storage_attestation","attest_author":"https://pith.science/pith/3FS5KGFDLWMT4ZNZRLFM3CCO4I/action/author_attestation","sign_citation":"https://pith.science/pith/3FS5KGFDLWMT4ZNZRLFM3CCO4I/action/citation_signature","submit_replication":"https://pith.science/pith/3FS5KGFDLWMT4ZNZRLFM3CCO4I/action/replication_record"}},"created_at":"2026-07-05T02:18:23.634388+00:00","updated_at":"2026-07-05T02:18:23.634388+00:00"}