{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:SZHIKSAJQCQSOMDS6DG2YH5CYR","short_pith_number":"pith:SZHIKSAJ","schema_version":"1.0","canonical_sha256":"964e85480980a1273072f0cdac1fa2c4687fc6d57e75d1f51aadfed83c0797c3","source":{"kind":"arxiv","id":"2410.14606","version":2},"attestation_state":"computed","paper":{"title":"Streaming Deep Reinforcement Learning Finally Works","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"A. Rupam Mahmood, Gautham Vasan, Mohamed Elsayed","submitted_at":"2024-10-18T17:00:29Z","abstract_excerpt":"Natural intelligence processes experience as a continuous stream, sensing, acting, and learning moment-by-moment in real time. Streaming learning, the modus operandi of classic reinforcement learning (RL) algorithms like Q-learning and TD, mimics natural learning by using the most recent sample without storing it. This approach is also ideal for resource-constrained, communication-limited, and privacy-sensitive applications. However, in deep RL, learners almost always use batch updates and replay buffers, making them computationally expensive and incompatible with streaming learning. Although "},"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":"2410.14606","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-18T17:00:29Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"fc20dc22f847c194fcd1a7e68d65b3b8d96a2d8c4a6dc163b34ff05c26f3c2ea","abstract_canon_sha256":"5a6cdda6afc66a7df532a0d4fa7565f8b9ccee7abff8bb55f64dbf08c3b75612"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:45:15.519166Z","signature_b64":"Ii7xeV69E2AgUP4et030WBigw/Xh7We/GiU56JTudP+to48qP71r5MA5bFC8mHjZRRBLjq9WVly02foNl6iJAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"964e85480980a1273072f0cdac1fa2c4687fc6d57e75d1f51aadfed83c0797c3","last_reissued_at":"2026-07-05T09:45:15.518698Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:45:15.518698Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Streaming Deep Reinforcement Learning Finally Works","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"A. Rupam Mahmood, Gautham Vasan, Mohamed Elsayed","submitted_at":"2024-10-18T17:00:29Z","abstract_excerpt":"Natural intelligence processes experience as a continuous stream, sensing, acting, and learning moment-by-moment in real time. Streaming learning, the modus operandi of classic reinforcement learning (RL) algorithms like Q-learning and TD, mimics natural learning by using the most recent sample without storing it. This approach is also ideal for resource-constrained, communication-limited, and privacy-sensitive applications. However, in deep RL, learners almost always use batch updates and replay buffers, making them computationally expensive and incompatible with streaming learning. Although "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.14606","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/2410.14606/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":"2410.14606","created_at":"2026-07-05T09:45:15.518752+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.14606v2","created_at":"2026-07-05T09:45:15.518752+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.14606","created_at":"2026-07-05T09:45:15.518752+00:00"},{"alias_kind":"pith_short_12","alias_value":"SZHIKSAJQCQS","created_at":"2026-07-05T09:45:15.518752+00:00"},{"alias_kind":"pith_short_16","alias_value":"SZHIKSAJQCQSOMDS","created_at":"2026-07-05T09:45:15.518752+00:00"},{"alias_kind":"pith_short_8","alias_value":"SZHIKSAJ","created_at":"2026-07-05T09:45:15.518752+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2410.17517","citing_title":"The Hive Mind is a Single Reinforcement Learning Agent","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19033","citing_title":"Intentional Updates for Streaming Reinforcement Learning","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2604.07945","citing_title":"Incremental Residual Reinforcement Learning Toward Real-World Learning for Social Navigation","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06764","citing_title":"Revisiting Adam for Streaming Reinforcement Learning","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04368","citing_title":"Extending Differential Temporal Difference Methods for Episodic Problems","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SZHIKSAJQCQSOMDS6DG2YH5CYR","json":"https://pith.science/pith/SZHIKSAJQCQSOMDS6DG2YH5CYR.json","graph_json":"https://pith.science/api/pith-number/SZHIKSAJQCQSOMDS6DG2YH5CYR/graph.json","events_json":"https://pith.science/api/pith-number/SZHIKSAJQCQSOMDS6DG2YH5CYR/events.json","paper":"https://pith.science/paper/SZHIKSAJ"},"agent_actions":{"view_html":"https://pith.science/pith/SZHIKSAJQCQSOMDS6DG2YH5CYR","download_json":"https://pith.science/pith/SZHIKSAJQCQSOMDS6DG2YH5CYR.json","view_paper":"https://pith.science/paper/SZHIKSAJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.14606&json=true","fetch_graph":"https://pith.science/api/pith-number/SZHIKSAJQCQSOMDS6DG2YH5CYR/graph.json","fetch_events":"https://pith.science/api/pith-number/SZHIKSAJQCQSOMDS6DG2YH5CYR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SZHIKSAJQCQSOMDS6DG2YH5CYR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SZHIKSAJQCQSOMDS6DG2YH5CYR/action/storage_attestation","attest_author":"https://pith.science/pith/SZHIKSAJQCQSOMDS6DG2YH5CYR/action/author_attestation","sign_citation":"https://pith.science/pith/SZHIKSAJQCQSOMDS6DG2YH5CYR/action/citation_signature","submit_replication":"https://pith.science/pith/SZHIKSAJQCQSOMDS6DG2YH5CYR/action/replication_record"}},"created_at":"2026-07-05T09:45:15.518752+00:00","updated_at":"2026-07-05T09:45:15.518752+00:00"}