{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:M7IMRYVCUEYIPXPZDLFLXERL2J","short_pith_number":"pith:M7IMRYVC","schema_version":"1.0","canonical_sha256":"67d0c8e2a2a13087ddf91acabb922bd253eaa840563c0dd66dd8d6b591e661ca","source":{"kind":"arxiv","id":"2106.07203","version":2},"attestation_state":"computed","paper":{"title":"Online Sub-Sampling for Reinforcement Learning with General Function Approximation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Dingwen Kong, Lin F. Yang, Ruosong Wang, Ruslan Salakhutdinov","submitted_at":"2021-06-14T07:36:25Z","abstract_excerpt":"Most of the existing works for reinforcement learning (RL) with general function approximation (FA) focus on understanding the statistical complexity or regret bounds. However, the computation complexity of such approaches is far from being understood -- indeed, a simple optimization problem over the function class might be as well intractable. In this paper, we tackle this problem by establishing an efficient online sub-sampling framework that measures the information gain of data points collected by an RL algorithm and uses the measurement to guide exploration. For a value-based method with "},"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":"2106.07203","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-06-14T07:36:25Z","cross_cats_sorted":["cs.AI","math.OC","stat.ML"],"title_canon_sha256":"24d8511a55eb2441448be85e66cfa60f93762f3d9d13c81db09dc86d06884790","abstract_canon_sha256":"66d4fcb99ea157bc4f24d8f837fe3a4988d863b04b80599dee4b0a1b67f813d8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:01:55.229174Z","signature_b64":"eq/MYZSFga/zFm4Wx1g71i2XDJwK53frEeqFFwBmt6NNmG+b10MmqUddx2TEjL5I5CgJI4KiGbYUsc0OHl/RAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"67d0c8e2a2a13087ddf91acabb922bd253eaa840563c0dd66dd8d6b591e661ca","last_reissued_at":"2026-07-05T06:01:55.228682Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:01:55.228682Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Online Sub-Sampling for Reinforcement Learning with General Function Approximation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Dingwen Kong, Lin F. Yang, Ruosong Wang, Ruslan Salakhutdinov","submitted_at":"2021-06-14T07:36:25Z","abstract_excerpt":"Most of the existing works for reinforcement learning (RL) with general function approximation (FA) focus on understanding the statistical complexity or regret bounds. However, the computation complexity of such approaches is far from being understood -- indeed, a simple optimization problem over the function class might be as well intractable. In this paper, we tackle this problem by establishing an efficient online sub-sampling framework that measures the information gain of data points collected by an RL algorithm and uses the measurement to guide exploration. For a value-based method with "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.07203","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/2106.07203/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":"2106.07203","created_at":"2026-07-05T06:01:55.228741+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.07203v2","created_at":"2026-07-05T06:01:55.228741+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.07203","created_at":"2026-07-05T06:01:55.228741+00:00"},{"alias_kind":"pith_short_12","alias_value":"M7IMRYVCUEYI","created_at":"2026-07-05T06:01:55.228741+00:00"},{"alias_kind":"pith_short_16","alias_value":"M7IMRYVCUEYIPXPZ","created_at":"2026-07-05T06:01:55.228741+00:00"},{"alias_kind":"pith_short_8","alias_value":"M7IMRYVC","created_at":"2026-07-05T06:01:55.228741+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.14821","citing_title":"Sample and Computationally Efficient Continuous-Time Reinforcement Learning with General Function Approximation","ref_index":25,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/M7IMRYVCUEYIPXPZDLFLXERL2J","json":"https://pith.science/pith/M7IMRYVCUEYIPXPZDLFLXERL2J.json","graph_json":"https://pith.science/api/pith-number/M7IMRYVCUEYIPXPZDLFLXERL2J/graph.json","events_json":"https://pith.science/api/pith-number/M7IMRYVCUEYIPXPZDLFLXERL2J/events.json","paper":"https://pith.science/paper/M7IMRYVC"},"agent_actions":{"view_html":"https://pith.science/pith/M7IMRYVCUEYIPXPZDLFLXERL2J","download_json":"https://pith.science/pith/M7IMRYVCUEYIPXPZDLFLXERL2J.json","view_paper":"https://pith.science/paper/M7IMRYVC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.07203&json=true","fetch_graph":"https://pith.science/api/pith-number/M7IMRYVCUEYIPXPZDLFLXERL2J/graph.json","fetch_events":"https://pith.science/api/pith-number/M7IMRYVCUEYIPXPZDLFLXERL2J/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/M7IMRYVCUEYIPXPZDLFLXERL2J/action/timestamp_anchor","attest_storage":"https://pith.science/pith/M7IMRYVCUEYIPXPZDLFLXERL2J/action/storage_attestation","attest_author":"https://pith.science/pith/M7IMRYVCUEYIPXPZDLFLXERL2J/action/author_attestation","sign_citation":"https://pith.science/pith/M7IMRYVCUEYIPXPZDLFLXERL2J/action/citation_signature","submit_replication":"https://pith.science/pith/M7IMRYVCUEYIPXPZDLFLXERL2J/action/replication_record"}},"created_at":"2026-07-05T06:01:55.228741+00:00","updated_at":"2026-07-05T06:01:55.228741+00:00"}