{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:5QYC3TDAOLIVAER26PR4YYCKNC","short_pith_number":"pith:5QYC3TDA","schema_version":"1.0","canonical_sha256":"ec302dcc6072d150123af3e3cc604a689c41066d89152cca8b5a816c178df683","source":{"kind":"arxiv","id":"2106.08053","version":1},"attestation_state":"computed","paper":{"title":"On the Power of Multitask Representation Learning in Linear MDP","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Gao Huang, Rui Lu, Simon S. Du","submitted_at":"2021-06-15T11:21:06Z","abstract_excerpt":"While multitask representation learning has become a popular approach in reinforcement learning (RL), theoretical understanding of why and when it works remains limited. This paper presents analyses for the statistical benefit of multitask representation learning in linear Markov Decision Process (MDP) under a generative model. In this paper, we consider an agent to learn a representation function $\\phi$ out of a function class $\\Phi$ from $T$ source tasks with $N$ data per task, and then use the learned $\\hat{\\phi}$ to reduce the required number of sample for a new task. We first discover a \\"},"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":"2106.08053","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-06-15T11:21:06Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ce24324536dd362bd6a7ebe5743e5b7279c60da4ff20e585986054158b5541e2","abstract_canon_sha256":"f5bf7ca320db32c2a6c6e16af06cea263d7c53670d0552f93245c13d3aface4e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:49:37.296871Z","signature_b64":"T2PI4JgX8hsmxkb9PyeocQQ5TXHaEGVecf6qg6bJFnXv4+p6l4akNtkhVfWGFWiF02aUU+gd09PFEBUpffW+AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ec302dcc6072d150123af3e3cc604a689c41066d89152cca8b5a816c178df683","last_reissued_at":"2026-07-05T02:49:37.296471Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:49:37.296471Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On the Power of Multitask Representation Learning in Linear MDP","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Gao Huang, Rui Lu, Simon S. Du","submitted_at":"2021-06-15T11:21:06Z","abstract_excerpt":"While multitask representation learning has become a popular approach in reinforcement learning (RL), theoretical understanding of why and when it works remains limited. This paper presents analyses for the statistical benefit of multitask representation learning in linear Markov Decision Process (MDP) under a generative model. In this paper, we consider an agent to learn a representation function $\\phi$ out of a function class $\\Phi$ from $T$ source tasks with $N$ data per task, and then use the learned $\\hat{\\phi}$ to reduce the required number of sample for a new task. We first discover a \\"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.08053","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/2106.08053/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":"2106.08053","created_at":"2026-07-05T02:49:37.296527+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.08053v1","created_at":"2026-07-05T02:49:37.296527+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.08053","created_at":"2026-07-05T02:49:37.296527+00:00"},{"alias_kind":"pith_short_12","alias_value":"5QYC3TDAOLIV","created_at":"2026-07-05T02:49:37.296527+00:00"},{"alias_kind":"pith_short_16","alias_value":"5QYC3TDAOLIVAER2","created_at":"2026-07-05T02:49:37.296527+00:00"},{"alias_kind":"pith_short_8","alias_value":"5QYC3TDA","created_at":"2026-07-05T02:49:37.296527+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.01242","citing_title":"Breaking the Computational Barrier: Provably Efficient Actor-Critic for Low-Rank MDPs","ref_index":75,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5QYC3TDAOLIVAER26PR4YYCKNC","json":"https://pith.science/pith/5QYC3TDAOLIVAER26PR4YYCKNC.json","graph_json":"https://pith.science/api/pith-number/5QYC3TDAOLIVAER26PR4YYCKNC/graph.json","events_json":"https://pith.science/api/pith-number/5QYC3TDAOLIVAER26PR4YYCKNC/events.json","paper":"https://pith.science/paper/5QYC3TDA"},"agent_actions":{"view_html":"https://pith.science/pith/5QYC3TDAOLIVAER26PR4YYCKNC","download_json":"https://pith.science/pith/5QYC3TDAOLIVAER26PR4YYCKNC.json","view_paper":"https://pith.science/paper/5QYC3TDA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.08053&json=true","fetch_graph":"https://pith.science/api/pith-number/5QYC3TDAOLIVAER26PR4YYCKNC/graph.json","fetch_events":"https://pith.science/api/pith-number/5QYC3TDAOLIVAER26PR4YYCKNC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5QYC3TDAOLIVAER26PR4YYCKNC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5QYC3TDAOLIVAER26PR4YYCKNC/action/storage_attestation","attest_author":"https://pith.science/pith/5QYC3TDAOLIVAER26PR4YYCKNC/action/author_attestation","sign_citation":"https://pith.science/pith/5QYC3TDAOLIVAER26PR4YYCKNC/action/citation_signature","submit_replication":"https://pith.science/pith/5QYC3TDAOLIVAER26PR4YYCKNC/action/replication_record"}},"created_at":"2026-07-05T02:49:37.296527+00:00","updated_at":"2026-07-05T02:49:37.296527+00:00"}