{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:YHYBG6LJFZME7J3APHLBDOK5G2","short_pith_number":"pith:YHYBG6LJ","canonical_record":{"source":{"id":"2112.15221","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2021-12-30T22:02:42Z","cross_cats_sorted":[],"title_canon_sha256":"e23dbf4d423b91c06431b91c6061dbb9efd8a8468754aa91009487322d3df244","abstract_canon_sha256":"a31730752b53afec99498c663fbed2400c5f78f5f2026879d8cbaf4485d2d87b"},"schema_version":"1.0"},"canonical_sha256":"c1f01379692e584fa76079d611b95d36aa31d88d21093e06ff7aae83e1ce9c40","source":{"kind":"arxiv","id":"2112.15221","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2112.15221","created_at":"2026-07-05T03:44:37Z"},{"alias_kind":"arxiv_version","alias_value":"2112.15221v1","created_at":"2026-07-05T03:44:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.15221","created_at":"2026-07-05T03:44:37Z"},{"alias_kind":"pith_short_12","alias_value":"YHYBG6LJFZME","created_at":"2026-07-05T03:44:37Z"},{"alias_kind":"pith_short_16","alias_value":"YHYBG6LJFZME7J3A","created_at":"2026-07-05T03:44:37Z"},{"alias_kind":"pith_short_8","alias_value":"YHYBG6LJ","created_at":"2026-07-05T03:44:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:YHYBG6LJFZME7J3APHLBDOK5G2","target":"record","payload":{"canonical_record":{"source":{"id":"2112.15221","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2021-12-30T22:02:42Z","cross_cats_sorted":[],"title_canon_sha256":"e23dbf4d423b91c06431b91c6061dbb9efd8a8468754aa91009487322d3df244","abstract_canon_sha256":"a31730752b53afec99498c663fbed2400c5f78f5f2026879d8cbaf4485d2d87b"},"schema_version":"1.0"},"canonical_sha256":"c1f01379692e584fa76079d611b95d36aa31d88d21093e06ff7aae83e1ce9c40","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:44:37.685942Z","signature_b64":"u3C17Uhlqbuy4Y1N2nm2uGBsHaqOAp8QkRqKjAfOARJD8HaEBXEfmfnBT14Dqyk/MZJbDYWBU4d4ksZGCNGuDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c1f01379692e584fa76079d611b95d36aa31d88d21093e06ff7aae83e1ce9c40","last_reissued_at":"2026-07-05T03:44:37.685568Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:44:37.685568Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2112.15221","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T03:44:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Poj8o18FSEmKsb/BkHdWkk/E8DU1BCOcxy43MVK/1jG4UM2QQoW3qiNMnHbyK18NkXuAkRpGnZGD6sAnWshkBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T06:54:28.889062Z"},"content_sha256":"05ea594c3c6b26f5d90ea4c14bdde69801440d0e9d71be0eeac30f8e4097b206","schema_version":"1.0","event_id":"sha256:05ea594c3c6b26f5d90ea4c14bdde69801440d0e9d71be0eeac30f8e4097b206"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:YHYBG6LJFZME7J3APHLBDOK5G2","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Constraint Sampling Reinforcement Learning: Incorporating Expertise For Faster Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"David Arbour, Emma Brunskill, Georgios Theocharous, Tong Mu","submitted_at":"2021-12-30T22:02:42Z","abstract_excerpt":"Online reinforcement learning (RL) algorithms are often difficult to deploy in complex human-facing applications as they may learn slowly and have poor early performance. To address this, we introduce a practical algorithm for incorporating human insight to speed learning. Our algorithm, Constraint Sampling Reinforcement Learning (CSRL), incorporates prior domain knowledge as constraints/restrictions on the RL policy. It takes in multiple potential policy constraints to maintain robustness to misspecification of individual constraints while leveraging helpful ones to learn quickly. Given a bas"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.15221","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/2112.15221/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":"2026-07-05T03:44:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Z32MnCO+yJoTOivjUXKlYcQ11lwww1TTPTkXgxLZI3YMU1QJLEIN7WrfPJadeQCo8+brsaHKDLUEetM9zV0HAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T06:54:28.889990Z"},"content_sha256":"f3e0094fd6e978c20eb3e1cfd8db52a40da10961d4256e8465ca2fe733f38b33","schema_version":"1.0","event_id":"sha256:f3e0094fd6e978c20eb3e1cfd8db52a40da10961d4256e8465ca2fe733f38b33"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YHYBG6LJFZME7J3APHLBDOK5G2/bundle.json","state_url":"https://pith.science/pith/YHYBG6LJFZME7J3APHLBDOK5G2/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YHYBG6LJFZME7J3APHLBDOK5G2/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":"2026-08-14T06:54:28Z","links":{"resolver":"https://pith.science/pith/YHYBG6LJFZME7J3APHLBDOK5G2","bundle":"https://pith.science/pith/YHYBG6LJFZME7J3APHLBDOK5G2/bundle.json","state":"https://pith.science/pith/YHYBG6LJFZME7J3APHLBDOK5G2/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YHYBG6LJFZME7J3APHLBDOK5G2/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:YHYBG6LJFZME7J3APHLBDOK5G2","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":"a31730752b53afec99498c663fbed2400c5f78f5f2026879d8cbaf4485d2d87b","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2021-12-30T22:02:42Z","title_canon_sha256":"e23dbf4d423b91c06431b91c6061dbb9efd8a8468754aa91009487322d3df244"},"schema_version":"1.0","source":{"id":"2112.15221","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2112.15221","created_at":"2026-07-05T03:44:37Z"},{"alias_kind":"arxiv_version","alias_value":"2112.15221v1","created_at":"2026-07-05T03:44:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.15221","created_at":"2026-07-05T03:44:37Z"},{"alias_kind":"pith_short_12","alias_value":"YHYBG6LJFZME","created_at":"2026-07-05T03:44:37Z"},{"alias_kind":"pith_short_16","alias_value":"YHYBG6LJFZME7J3A","created_at":"2026-07-05T03:44:37Z"},{"alias_kind":"pith_short_8","alias_value":"YHYBG6LJ","created_at":"2026-07-05T03:44:37Z"}],"graph_snapshots":[{"event_id":"sha256:f3e0094fd6e978c20eb3e1cfd8db52a40da10961d4256e8465ca2fe733f38b33","target":"graph","created_at":"2026-07-05T03:44:37Z","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/2112.15221/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Online reinforcement learning (RL) algorithms are often difficult to deploy in complex human-facing applications as they may learn slowly and have poor early performance. To address this, we introduce a practical algorithm for incorporating human insight to speed learning. Our algorithm, Constraint Sampling Reinforcement Learning (CSRL), incorporates prior domain knowledge as constraints/restrictions on the RL policy. It takes in multiple potential policy constraints to maintain robustness to misspecification of individual constraints while leveraging helpful ones to learn quickly. Given a bas","authors_text":"David Arbour, Emma Brunskill, Georgios Theocharous, Tong Mu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2021-12-30T22:02:42Z","title":"Constraint Sampling Reinforcement Learning: Incorporating Expertise For Faster Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.15221","kind":"arxiv","version":1},"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":"sha256:05ea594c3c6b26f5d90ea4c14bdde69801440d0e9d71be0eeac30f8e4097b206","target":"record","created_at":"2026-07-05T03:44:37Z","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":"a31730752b53afec99498c663fbed2400c5f78f5f2026879d8cbaf4485d2d87b","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2021-12-30T22:02:42Z","title_canon_sha256":"e23dbf4d423b91c06431b91c6061dbb9efd8a8468754aa91009487322d3df244"},"schema_version":"1.0","source":{"id":"2112.15221","kind":"arxiv","version":1}},"canonical_sha256":"c1f01379692e584fa76079d611b95d36aa31d88d21093e06ff7aae83e1ce9c40","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c1f01379692e584fa76079d611b95d36aa31d88d21093e06ff7aae83e1ce9c40","first_computed_at":"2026-07-05T03:44:37.685568Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:44:37.685568Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"u3C17Uhlqbuy4Y1N2nm2uGBsHaqOAp8QkRqKjAfOARJD8HaEBXEfmfnBT14Dqyk/MZJbDYWBU4d4ksZGCNGuDw==","signature_status":"signed_v1","signed_at":"2026-07-05T03:44:37.685942Z","signed_message":"canonical_sha256_bytes"},"source_id":"2112.15221","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:05ea594c3c6b26f5d90ea4c14bdde69801440d0e9d71be0eeac30f8e4097b206","sha256:f3e0094fd6e978c20eb3e1cfd8db52a40da10961d4256e8465ca2fe733f38b33"],"state_sha256":"dbf2143b100fb23f0c836c1a734173602613d623a7beb30a172e123c4ad9fa33"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RRaSGNwFJDj8FFEWNq/sgF6L7FosUm/1x9bwUcWucjOMP9yERBtrxI0eCEfHH9ctaJC750ucSF8gEofWrBXKBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T06:54:28.896486Z","bundle_sha256":"0c4c24b7a85282201af55421dcc30d4d2934bb59cf26d1fbdf1eeab69dbaebf4"}}