{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:YYYWHEI63AMWDUO2EQL7DYL6CL","short_pith_number":"pith:YYYWHEI6","schema_version":"1.0","canonical_sha256":"c63163911ed81961d1da2417f1e17e12c0992b06a4119c5edc4c527e0f5ab747","source":{"kind":"arxiv","id":"2401.17173","version":3},"attestation_state":"computed","paper":{"title":"Zero-Shot Reinforcement Learning via Function Encoders","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Amy Zhang, Tyler Ingebrand, Ufuk Topcu","submitted_at":"2024-01-30T17:04:47Z","abstract_excerpt":"Although reinforcement learning (RL) can solve many challenging sequential decision making problems, achieving zero-shot transfer across related tasks remains a challenge. The difficulty lies in finding a good representation for the current task so that the agent understands how it relates to previously seen tasks. To achieve zero-shot transfer, we introduce the function encoder, a representation learning algorithm which represents a function as a weighted combination of learned, non-linear basis functions. By using a function encoder to represent the reward function or the transition function"},"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":"2401.17173","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-01-30T17:04:47Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b519ba504cb3ca64f3c2db261ab8009f2beb10373ed24e583732cca82e3b1237","abstract_canon_sha256":"167f2813056490400937e0b8c78dd42012d10ca4a264b4e382cf99bb7d7ba71c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:36:23.429577Z","signature_b64":"ienrZLed1zXkiFkrHCZAZZP8QymiIvpSV7UQz5gBTSXBdaXBk0pPAhncwV9hkPR6lYPxvLTC70vlPsQEubigAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c63163911ed81961d1da2417f1e17e12c0992b06a4119c5edc4c527e0f5ab747","last_reissued_at":"2026-07-05T10:36:23.429066Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:36:23.429066Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Zero-Shot Reinforcement Learning via Function Encoders","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Amy Zhang, Tyler Ingebrand, Ufuk Topcu","submitted_at":"2024-01-30T17:04:47Z","abstract_excerpt":"Although reinforcement learning (RL) can solve many challenging sequential decision making problems, achieving zero-shot transfer across related tasks remains a challenge. The difficulty lies in finding a good representation for the current task so that the agent understands how it relates to previously seen tasks. To achieve zero-shot transfer, we introduce the function encoder, a representation learning algorithm which represents a function as a weighted combination of learned, non-linear basis functions. By using a function encoder to represent the reward function or the transition function"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.17173","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/2401.17173/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":"2401.17173","created_at":"2026-07-05T10:36:23.429126+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.17173v3","created_at":"2026-07-05T10:36:23.429126+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.17173","created_at":"2026-07-05T10:36:23.429126+00:00"},{"alias_kind":"pith_short_12","alias_value":"YYYWHEI63AMW","created_at":"2026-07-05T10:36:23.429126+00:00"},{"alias_kind":"pith_short_16","alias_value":"YYYWHEI63AMWDUO2","created_at":"2026-07-05T10:36:23.429126+00:00"},{"alias_kind":"pith_short_8","alias_value":"YYYWHEI6","created_at":"2026-07-05T10:36:23.429126+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.25496","citing_title":"Improving Zero-Shot Offline RL via Behavioral Task Sampling","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YYYWHEI63AMWDUO2EQL7DYL6CL","json":"https://pith.science/pith/YYYWHEI63AMWDUO2EQL7DYL6CL.json","graph_json":"https://pith.science/api/pith-number/YYYWHEI63AMWDUO2EQL7DYL6CL/graph.json","events_json":"https://pith.science/api/pith-number/YYYWHEI63AMWDUO2EQL7DYL6CL/events.json","paper":"https://pith.science/paper/YYYWHEI6"},"agent_actions":{"view_html":"https://pith.science/pith/YYYWHEI63AMWDUO2EQL7DYL6CL","download_json":"https://pith.science/pith/YYYWHEI63AMWDUO2EQL7DYL6CL.json","view_paper":"https://pith.science/paper/YYYWHEI6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.17173&json=true","fetch_graph":"https://pith.science/api/pith-number/YYYWHEI63AMWDUO2EQL7DYL6CL/graph.json","fetch_events":"https://pith.science/api/pith-number/YYYWHEI63AMWDUO2EQL7DYL6CL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YYYWHEI63AMWDUO2EQL7DYL6CL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YYYWHEI63AMWDUO2EQL7DYL6CL/action/storage_attestation","attest_author":"https://pith.science/pith/YYYWHEI63AMWDUO2EQL7DYL6CL/action/author_attestation","sign_citation":"https://pith.science/pith/YYYWHEI63AMWDUO2EQL7DYL6CL/action/citation_signature","submit_replication":"https://pith.science/pith/YYYWHEI63AMWDUO2EQL7DYL6CL/action/replication_record"}},"created_at":"2026-07-05T10:36:23.429126+00:00","updated_at":"2026-07-05T10:36:23.429126+00:00"}