{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:H5DKWJAFG6D7FBBIMWNAEWPNW7","short_pith_number":"pith:H5DKWJAF","schema_version":"1.0","canonical_sha256":"3f46ab24053787f28428659a0259edb7fda40404c4d193bce278effd4886a1ec","source":{"kind":"arxiv","id":"2206.04779","version":3},"attestation_state":"computed","paper":{"title":"Challenges and Opportunities in Offline Reinforcement Learning from Visual Observations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Cong Lu, Jack Parker-Holder, Michael A. Osborne, Philip J. Ball, Tim G. J. Rudner, Yee Whye Teh","submitted_at":"2022-06-09T22:08:47Z","abstract_excerpt":"Offline reinforcement learning has shown great promise in leveraging large pre-collected datasets for policy learning, allowing agents to forgo often-expensive online data collection. However, offline reinforcement learning from visual observations with continuous action spaces remains under-explored, with a limited understanding of the key challenges in this complex domain. In this paper, we establish simple baselines for continuous control in the visual domain and introduce a suite of benchmarking tasks for offline reinforcement learning from visual observations designed to better represent "},"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":"2206.04779","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-06-09T22:08:47Z","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"title_canon_sha256":"5c0a0df5e94452d1d5b6abeb214c5d1280286024c1f014b19cd1d84e007c070e","abstract_canon_sha256":"841ee9660083b9c55733b38d4ab9ab8a205dae563465fe4cd73c011d7fdf5e14"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:28:13.349050Z","signature_b64":"LM+lqUCzMSuEGoaCTr3+O1R5IQdx7t/9Dn51+iEj6kOc8hzMbmPG7LQYKbcsTKRFLpxpG9hNfMjq6qUJW5bUCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3f46ab24053787f28428659a0259edb7fda40404c4d193bce278effd4886a1ec","last_reissued_at":"2026-07-05T06:28:13.348571Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:28:13.348571Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Challenges and Opportunities in Offline Reinforcement Learning from Visual Observations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Cong Lu, Jack Parker-Holder, Michael A. Osborne, Philip J. Ball, Tim G. J. Rudner, Yee Whye Teh","submitted_at":"2022-06-09T22:08:47Z","abstract_excerpt":"Offline reinforcement learning has shown great promise in leveraging large pre-collected datasets for policy learning, allowing agents to forgo often-expensive online data collection. However, offline reinforcement learning from visual observations with continuous action spaces remains under-explored, with a limited understanding of the key challenges in this complex domain. In this paper, we establish simple baselines for continuous control in the visual domain and introduce a suite of benchmarking tasks for offline reinforcement learning from visual observations designed to better represent "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.04779","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/2206.04779/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":"2206.04779","created_at":"2026-07-05T06:28:13.348628+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.04779v3","created_at":"2026-07-05T06:28:13.348628+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.04779","created_at":"2026-07-05T06:28:13.348628+00:00"},{"alias_kind":"pith_short_12","alias_value":"H5DKWJAFG6D7","created_at":"2026-07-05T06:28:13.348628+00:00"},{"alias_kind":"pith_short_16","alias_value":"H5DKWJAFG6D7FBBI","created_at":"2026-07-05T06:28:13.348628+00:00"},{"alias_kind":"pith_short_8","alias_value":"H5DKWJAF","created_at":"2026-07-05T06:28:13.348628+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2602.07399","citing_title":"VGAS: Value-Guided Action-Chunk Selection for Few-Shot Vision-Language-Action Adaptation","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12334","citing_title":"Reinforcing VLAs in Task-Agnostic World Models","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12334","citing_title":"Reinforcing VLAs in Task-Agnostic World Models","ref_index":23,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/H5DKWJAFG6D7FBBIMWNAEWPNW7","json":"https://pith.science/pith/H5DKWJAFG6D7FBBIMWNAEWPNW7.json","graph_json":"https://pith.science/api/pith-number/H5DKWJAFG6D7FBBIMWNAEWPNW7/graph.json","events_json":"https://pith.science/api/pith-number/H5DKWJAFG6D7FBBIMWNAEWPNW7/events.json","paper":"https://pith.science/paper/H5DKWJAF"},"agent_actions":{"view_html":"https://pith.science/pith/H5DKWJAFG6D7FBBIMWNAEWPNW7","download_json":"https://pith.science/pith/H5DKWJAFG6D7FBBIMWNAEWPNW7.json","view_paper":"https://pith.science/paper/H5DKWJAF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.04779&json=true","fetch_graph":"https://pith.science/api/pith-number/H5DKWJAFG6D7FBBIMWNAEWPNW7/graph.json","fetch_events":"https://pith.science/api/pith-number/H5DKWJAFG6D7FBBIMWNAEWPNW7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/H5DKWJAFG6D7FBBIMWNAEWPNW7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/H5DKWJAFG6D7FBBIMWNAEWPNW7/action/storage_attestation","attest_author":"https://pith.science/pith/H5DKWJAFG6D7FBBIMWNAEWPNW7/action/author_attestation","sign_citation":"https://pith.science/pith/H5DKWJAFG6D7FBBIMWNAEWPNW7/action/citation_signature","submit_replication":"https://pith.science/pith/H5DKWJAFG6D7FBBIMWNAEWPNW7/action/replication_record"}},"created_at":"2026-07-05T06:28:13.348628+00:00","updated_at":"2026-07-05T06:28:13.348628+00:00"}