{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:VH637U7TSHODGTZCFMS3M4ZSL6","short_pith_number":"pith:VH637U7T","schema_version":"1.0","canonical_sha256":"a9fdbfd3f391dc334f222b25b673325f98f2accce5d78b2e1d73664a6de5383b","source":{"kind":"arxiv","id":"2402.03141","version":2},"attestation_state":"computed","paper":{"title":"Boosting Reinforcement Learning with Strongly Delayed Feedback Through Auxiliary Short Delays","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.SY","eess.SY"],"primary_cat":"cs.LG","authors_text":"Chao Huang, Chen Lv, Chung-Wei Lin, J\\\"urgen Schmidhuber, Qingyuan Wu, Qi Zhu, Simon Sinong Zhan, Yixuan Wang, Yuhui Wang","submitted_at":"2024-02-05T16:11:03Z","abstract_excerpt":"Reinforcement learning (RL) is challenging in the common case of delays between events and their sensory perceptions. State-of-the-art (SOTA) state augmentation techniques either suffer from state space explosion or performance degeneration in stochastic environments. To address these challenges, we present a novel Auxiliary-Delayed Reinforcement Learning (AD-RL) method that leverages auxiliary tasks involving short delays to accelerate RL with long delays, without compromising performance in stochastic environments. Specifically, AD-RL learns a value function for short delays and uses bootstr"},"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":"2402.03141","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-05T16:11:03Z","cross_cats_sorted":["cs.AI","cs.SY","eess.SY"],"title_canon_sha256":"7909ffc9f893abc546cd6c7a014f15bbd51bde7f801c4b7f5aaa2d7f93bb82cd","abstract_canon_sha256":"52c736dce7e4e761ba80f141cb16266cd2353ae5217591d8d4b876bffaa1eba9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:28:05.528183Z","signature_b64":"agNYH4gBXIeQdruRwk09woZRNKIzWlBKRSQYd8Utr5jbhrPu3UzVywjoUPexk+yOp1IddywsKDv+KS8PVh6lCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a9fdbfd3f391dc334f222b25b673325f98f2accce5d78b2e1d73664a6de5383b","last_reissued_at":"2026-07-05T08:28:05.527759Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:28:05.527759Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Boosting Reinforcement Learning with Strongly Delayed Feedback Through Auxiliary Short Delays","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.SY","eess.SY"],"primary_cat":"cs.LG","authors_text":"Chao Huang, Chen Lv, Chung-Wei Lin, J\\\"urgen Schmidhuber, Qingyuan Wu, Qi Zhu, Simon Sinong Zhan, Yixuan Wang, Yuhui Wang","submitted_at":"2024-02-05T16:11:03Z","abstract_excerpt":"Reinforcement learning (RL) is challenging in the common case of delays between events and their sensory perceptions. State-of-the-art (SOTA) state augmentation techniques either suffer from state space explosion or performance degeneration in stochastic environments. To address these challenges, we present a novel Auxiliary-Delayed Reinforcement Learning (AD-RL) method that leverages auxiliary tasks involving short delays to accelerate RL with long delays, without compromising performance in stochastic environments. Specifically, AD-RL learns a value function for short delays and uses bootstr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.03141","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/2402.03141/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":"2402.03141","created_at":"2026-07-05T08:28:05.527814+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.03141v2","created_at":"2026-07-05T08:28:05.527814+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.03141","created_at":"2026-07-05T08:28:05.527814+00:00"},{"alias_kind":"pith_short_12","alias_value":"VH637U7TSHOD","created_at":"2026-07-05T08:28:05.527814+00:00"},{"alias_kind":"pith_short_16","alias_value":"VH637U7TSHODGTZC","created_at":"2026-07-05T08:28:05.527814+00:00"},{"alias_kind":"pith_short_8","alias_value":"VH637U7T","created_at":"2026-07-05T08:28:05.527814+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.20869","citing_title":"Model-Based Reinforcement Learning under Random Observation Delays","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04345","citing_title":"Structural Equivalence and Learning Dynamics in Delayed MARL","ref_index":23,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VH637U7TSHODGTZCFMS3M4ZSL6","json":"https://pith.science/pith/VH637U7TSHODGTZCFMS3M4ZSL6.json","graph_json":"https://pith.science/api/pith-number/VH637U7TSHODGTZCFMS3M4ZSL6/graph.json","events_json":"https://pith.science/api/pith-number/VH637U7TSHODGTZCFMS3M4ZSL6/events.json","paper":"https://pith.science/paper/VH637U7T"},"agent_actions":{"view_html":"https://pith.science/pith/VH637U7TSHODGTZCFMS3M4ZSL6","download_json":"https://pith.science/pith/VH637U7TSHODGTZCFMS3M4ZSL6.json","view_paper":"https://pith.science/paper/VH637U7T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.03141&json=true","fetch_graph":"https://pith.science/api/pith-number/VH637U7TSHODGTZCFMS3M4ZSL6/graph.json","fetch_events":"https://pith.science/api/pith-number/VH637U7TSHODGTZCFMS3M4ZSL6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VH637U7TSHODGTZCFMS3M4ZSL6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VH637U7TSHODGTZCFMS3M4ZSL6/action/storage_attestation","attest_author":"https://pith.science/pith/VH637U7TSHODGTZCFMS3M4ZSL6/action/author_attestation","sign_citation":"https://pith.science/pith/VH637U7TSHODGTZCFMS3M4ZSL6/action/citation_signature","submit_replication":"https://pith.science/pith/VH637U7TSHODGTZCFMS3M4ZSL6/action/replication_record"}},"created_at":"2026-07-05T08:28:05.527814+00:00","updated_at":"2026-07-05T08:28:05.527814+00:00"}