{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:BL5YMUE3RPP6JPYJGOSZV2BVTT","short_pith_number":"pith:BL5YMUE3","canonical_record":{"source":{"id":"2204.12581","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-04-26T20:42:14Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9c2e4a89f6ae20e89492fa59b17bfb686c8f77ecf91b5c6870e8033a0038c57c","abstract_canon_sha256":"fecd266d986b3343f21ed8ec64a833f9f3d9f80658c5bb4e6af8a98258e2ac91"},"schema_version":"1.0"},"canonical_sha256":"0afb86509b8bdfe4bf0933a59ae8359cca64a4755aab9fce0bfab0a5909ddf16","source":{"kind":"arxiv","id":"2204.12581","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2204.12581","created_at":"2026-07-05T05:05:05Z"},{"alias_kind":"arxiv_version","alias_value":"2204.12581v3","created_at":"2026-07-05T05:05:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2204.12581","created_at":"2026-07-05T05:05:05Z"},{"alias_kind":"pith_short_12","alias_value":"BL5YMUE3RPP6","created_at":"2026-07-05T05:05:05Z"},{"alias_kind":"pith_short_16","alias_value":"BL5YMUE3RPP6JPYJ","created_at":"2026-07-05T05:05:05Z"},{"alias_kind":"pith_short_8","alias_value":"BL5YMUE3","created_at":"2026-07-05T05:05:05Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:BL5YMUE3RPP6JPYJGOSZV2BVTT","target":"record","payload":{"canonical_record":{"source":{"id":"2204.12581","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-04-26T20:42:14Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9c2e4a89f6ae20e89492fa59b17bfb686c8f77ecf91b5c6870e8033a0038c57c","abstract_canon_sha256":"fecd266d986b3343f21ed8ec64a833f9f3d9f80658c5bb4e6af8a98258e2ac91"},"schema_version":"1.0"},"canonical_sha256":"0afb86509b8bdfe4bf0933a59ae8359cca64a4755aab9fce0bfab0a5909ddf16","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:05:05.074246Z","signature_b64":"vTBGqgevJTYyod7FMmHAOuI/iBEoYI2IHF6vzul8naom7qbMNR7tzCxHJYiYP4zCL3ohX0bOALV+vpQvy0XNDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0afb86509b8bdfe4bf0933a59ae8359cca64a4755aab9fce0bfab0a5909ddf16","last_reissued_at":"2026-07-05T05:05:05.073825Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:05:05.073825Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2204.12581","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:05:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"muvxfDRw/9qdRraiD+TUDphGbbDP8Q8tgSemLnAEoqApu0f58e8q/aidC2wqCamgUGjhr5l5RTwY3LjPiaqODg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T00:50:08.390814Z"},"content_sha256":"3e0ecc414f28c478c4f0255680a53f02aba4f81133c4c8f30cff403d69a24141","schema_version":"1.0","event_id":"sha256:3e0ecc414f28c478c4f0255680a53f02aba4f81133c4c8f30cff403d69a24141"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:BL5YMUE3RPP6JPYJGOSZV2BVTT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"RAMBO-RL: Robust Adversarial Model-Based Offline Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bruno Lacerda, Marc Rigter, Nick Hawes","submitted_at":"2022-04-26T20:42:14Z","abstract_excerpt":"Offline reinforcement learning (RL) aims to find performant policies from logged data without further environment interaction. Model-based algorithms, which learn a model of the environment from the dataset and perform conservative policy optimisation within that model, have emerged as a promising approach to this problem. In this work, we present Robust Adversarial Model-Based Offline RL (RAMBO), a novel approach to model-based offline RL. We formulate the problem as a two-player zero sum game against an adversarial environment model. The model is trained to minimise the value function while "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2204.12581","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/2204.12581/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:05:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Nk3JGozEMxfhAgXDhrNpKMqovgQk0DVTY2XYy66mYgG27seffHFvNro+pTSjMYwlCoRDZq2+6X3sxLTanmPuDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T00:50:08.391307Z"},"content_sha256":"1ad9e1733e89c37f41e202e64e5f70c782e303ad9bc2da49b8190d749a89f3e0","schema_version":"1.0","event_id":"sha256:1ad9e1733e89c37f41e202e64e5f70c782e303ad9bc2da49b8190d749a89f3e0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BL5YMUE3RPP6JPYJGOSZV2BVTT/bundle.json","state_url":"https://pith.science/pith/BL5YMUE3RPP6JPYJGOSZV2BVTT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BL5YMUE3RPP6JPYJGOSZV2BVTT/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-10T00:50:08Z","links":{"resolver":"https://pith.science/pith/BL5YMUE3RPP6JPYJGOSZV2BVTT","bundle":"https://pith.science/pith/BL5YMUE3RPP6JPYJGOSZV2BVTT/bundle.json","state":"https://pith.science/pith/BL5YMUE3RPP6JPYJGOSZV2BVTT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BL5YMUE3RPP6JPYJGOSZV2BVTT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:BL5YMUE3RPP6JPYJGOSZV2BVTT","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"fecd266d986b3343f21ed8ec64a833f9f3d9f80658c5bb4e6af8a98258e2ac91","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-04-26T20:42:14Z","title_canon_sha256":"9c2e4a89f6ae20e89492fa59b17bfb686c8f77ecf91b5c6870e8033a0038c57c"},"schema_version":"1.0","source":{"id":"2204.12581","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2204.12581","created_at":"2026-07-05T05:05:05Z"},{"alias_kind":"arxiv_version","alias_value":"2204.12581v3","created_at":"2026-07-05T05:05:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2204.12581","created_at":"2026-07-05T05:05:05Z"},{"alias_kind":"pith_short_12","alias_value":"BL5YMUE3RPP6","created_at":"2026-07-05T05:05:05Z"},{"alias_kind":"pith_short_16","alias_value":"BL5YMUE3RPP6JPYJ","created_at":"2026-07-05T05:05:05Z"},{"alias_kind":"pith_short_8","alias_value":"BL5YMUE3","created_at":"2026-07-05T05:05:05Z"}],"graph_snapshots":[{"event_id":"sha256:1ad9e1733e89c37f41e202e64e5f70c782e303ad9bc2da49b8190d749a89f3e0","target":"graph","created_at":"2026-07-05T05:05:05Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2204.12581/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Offline reinforcement learning (RL) aims to find performant policies from logged data without further environment interaction. Model-based algorithms, which learn a model of the environment from the dataset and perform conservative policy optimisation within that model, have emerged as a promising approach to this problem. In this work, we present Robust Adversarial Model-Based Offline RL (RAMBO), a novel approach to model-based offline RL. We formulate the problem as a two-player zero sum game against an adversarial environment model. The model is trained to minimise the value function while ","authors_text":"Bruno Lacerda, Marc Rigter, Nick Hawes","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-04-26T20:42:14Z","title":"RAMBO-RL: Robust Adversarial Model-Based Offline Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2204.12581","kind":"arxiv","version":3},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:3e0ecc414f28c478c4f0255680a53f02aba4f81133c4c8f30cff403d69a24141","target":"record","created_at":"2026-07-05T05:05:05Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"fecd266d986b3343f21ed8ec64a833f9f3d9f80658c5bb4e6af8a98258e2ac91","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-04-26T20:42:14Z","title_canon_sha256":"9c2e4a89f6ae20e89492fa59b17bfb686c8f77ecf91b5c6870e8033a0038c57c"},"schema_version":"1.0","source":{"id":"2204.12581","kind":"arxiv","version":3}},"canonical_sha256":"0afb86509b8bdfe4bf0933a59ae8359cca64a4755aab9fce0bfab0a5909ddf16","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0afb86509b8bdfe4bf0933a59ae8359cca64a4755aab9fce0bfab0a5909ddf16","first_computed_at":"2026-07-05T05:05:05.073825Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:05:05.073825Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vTBGqgevJTYyod7FMmHAOuI/iBEoYI2IHF6vzul8naom7qbMNR7tzCxHJYiYP4zCL3ohX0bOALV+vpQvy0XNDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T05:05:05.074246Z","signed_message":"canonical_sha256_bytes"},"source_id":"2204.12581","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3e0ecc414f28c478c4f0255680a53f02aba4f81133c4c8f30cff403d69a24141","sha256:1ad9e1733e89c37f41e202e64e5f70c782e303ad9bc2da49b8190d749a89f3e0"],"state_sha256":"a96561a78f94cd308cb73a65629852f917633cff59e2648bace5798e39e8a2ce"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1NOVIjE2Q8w+0DDh1stseOxg1jihMHL6Ul5DqAOvUys7x/flfWCXBS7JPrwdApdXDqfqs94Ya0J/Tf+W/6MKDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T00:50:08.394646Z","bundle_sha256":"c22c120d368b7db84633a53bb3de9221a46a89f6c02defa6f69e4a3f646ccd87"}}