{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:LPEWQCN4Z46LYWFD6KFLPFDEIL","short_pith_number":"pith:LPEWQCN4","schema_version":"1.0","canonical_sha256":"5bc96809bccf3cbc58a3f28ab7946442d1e93ee116acde6a975ac9e3ed246fb8","source":{"kind":"arxiv","id":"2405.03064","version":3},"attestation_state":"computed","paper":{"title":"RICE: Breaking Through the Training Bottlenecks of Reinforcement Learning with Explanation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.LG","authors_text":"Gang Wang, Jiahao Yu, Sabrina Yang, Xian Wu, Xinyu Xing, Zelei Cheng","submitted_at":"2024-05-05T22:06:42Z","abstract_excerpt":"Deep reinforcement learning (DRL) is playing an increasingly important role in real-world applications. However, obtaining an optimally performing DRL agent for complex tasks, especially with sparse rewards, remains a significant challenge. The training of a DRL agent can be often trapped in a bottleneck without further progress. In this paper, we propose RICE, an innovative refining scheme for reinforcement learning that incorporates explanation methods to break through the training bottlenecks. The high-level idea of RICE is to construct a new initial state distribution that combines both th"},"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":"2405.03064","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-05-05T22:06:42Z","cross_cats_sorted":["cs.AI","cs.CR"],"title_canon_sha256":"792a93d35ee04f244022584dfdc0c96f36bbb02f59388e6402474a98cd1b2613","abstract_canon_sha256":"41e968d3d113128c8050ff650a63453afba39e7c3aa9c791836a7ed0f247fde7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:28:09.800038Z","signature_b64":"iktd7exmHOUjdlJwzvQK6cif46jiheLnfoax+kWNsVRZOWNfViLtJ+yuC6uUiv0GIoiteZomLtmfC9ocxHoHCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5bc96809bccf3cbc58a3f28ab7946442d1e93ee116acde6a975ac9e3ed246fb8","last_reissued_at":"2026-07-05T08:28:09.799579Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:28:09.799579Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RICE: Breaking Through the Training Bottlenecks of Reinforcement Learning with Explanation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.LG","authors_text":"Gang Wang, Jiahao Yu, Sabrina Yang, Xian Wu, Xinyu Xing, Zelei Cheng","submitted_at":"2024-05-05T22:06:42Z","abstract_excerpt":"Deep reinforcement learning (DRL) is playing an increasingly important role in real-world applications. However, obtaining an optimally performing DRL agent for complex tasks, especially with sparse rewards, remains a significant challenge. The training of a DRL agent can be often trapped in a bottleneck without further progress. In this paper, we propose RICE, an innovative refining scheme for reinforcement learning that incorporates explanation methods to break through the training bottlenecks. The high-level idea of RICE is to construct a new initial state distribution that combines both th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.03064","kind":"arxiv","version":3},"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/2405.03064/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":"2405.03064","created_at":"2026-07-05T08:28:09.799636+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.03064v3","created_at":"2026-07-05T08:28:09.799636+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.03064","created_at":"2026-07-05T08:28:09.799636+00:00"},{"alias_kind":"pith_short_12","alias_value":"LPEWQCN4Z46L","created_at":"2026-07-05T08:28:09.799636+00:00"},{"alias_kind":"pith_short_16","alias_value":"LPEWQCN4Z46LYWFD","created_at":"2026-07-05T08:28:09.799636+00:00"},{"alias_kind":"pith_short_8","alias_value":"LPEWQCN4","created_at":"2026-07-05T08:28:09.799636+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.20671","citing_title":"LLM-Guided Reinforcement Learning: Addressing Training Bottlenecks through Policy Modulation","ref_index":7,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LPEWQCN4Z46LYWFD6KFLPFDEIL","json":"https://pith.science/pith/LPEWQCN4Z46LYWFD6KFLPFDEIL.json","graph_json":"https://pith.science/api/pith-number/LPEWQCN4Z46LYWFD6KFLPFDEIL/graph.json","events_json":"https://pith.science/api/pith-number/LPEWQCN4Z46LYWFD6KFLPFDEIL/events.json","paper":"https://pith.science/paper/LPEWQCN4"},"agent_actions":{"view_html":"https://pith.science/pith/LPEWQCN4Z46LYWFD6KFLPFDEIL","download_json":"https://pith.science/pith/LPEWQCN4Z46LYWFD6KFLPFDEIL.json","view_paper":"https://pith.science/paper/LPEWQCN4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.03064&json=true","fetch_graph":"https://pith.science/api/pith-number/LPEWQCN4Z46LYWFD6KFLPFDEIL/graph.json","fetch_events":"https://pith.science/api/pith-number/LPEWQCN4Z46LYWFD6KFLPFDEIL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LPEWQCN4Z46LYWFD6KFLPFDEIL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LPEWQCN4Z46LYWFD6KFLPFDEIL/action/storage_attestation","attest_author":"https://pith.science/pith/LPEWQCN4Z46LYWFD6KFLPFDEIL/action/author_attestation","sign_citation":"https://pith.science/pith/LPEWQCN4Z46LYWFD6KFLPFDEIL/action/citation_signature","submit_replication":"https://pith.science/pith/LPEWQCN4Z46LYWFD6KFLPFDEIL/action/replication_record"}},"created_at":"2026-07-05T08:28:09.799636+00:00","updated_at":"2026-07-05T08:28:09.799636+00:00"}