{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:YOUXTHHEJ72NQ7S4FLQ7EGA33Z","short_pith_number":"pith:YOUXTHHE","canonical_record":{"source":{"id":"1805.08313","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-05-21T22:40:07Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"2e7755388a1c3fbdc166926bb789eeca5ac4dbe03b5b770bfd9fbe9e9d7a6d1d","abstract_canon_sha256":"1875a140d1ebf49f3c3ee82cd15d02449aff3f750e4fa078cbcdbfb0d71a418a"},"schema_version":"1.0"},"canonical_sha256":"c3a9799ce44ff4d87e5c2ae1f2181bde6770a205d6e2150766e9d859c880eb61","source":{"kind":"arxiv","id":"1805.08313","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1805.08313","created_at":"2026-05-18T00:00:13Z"},{"alias_kind":"arxiv_version","alias_value":"1805.08313v2","created_at":"2026-05-18T00:00:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1805.08313","created_at":"2026-05-18T00:00:13Z"},{"alias_kind":"pith_short_12","alias_value":"YOUXTHHEJ72N","created_at":"2026-05-18T12:33:04Z"},{"alias_kind":"pith_short_16","alias_value":"YOUXTHHEJ72NQ7S4","created_at":"2026-05-18T12:33:04Z"},{"alias_kind":"pith_short_8","alias_value":"YOUXTHHE","created_at":"2026-05-18T12:33:04Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:YOUXTHHEJ72NQ7S4FLQ7EGA33Z","target":"record","payload":{"canonical_record":{"source":{"id":"1805.08313","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-05-21T22:40:07Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"2e7755388a1c3fbdc166926bb789eeca5ac4dbe03b5b770bfd9fbe9e9d7a6d1d","abstract_canon_sha256":"1875a140d1ebf49f3c3ee82cd15d02449aff3f750e4fa078cbcdbfb0d71a418a"},"schema_version":"1.0"},"canonical_sha256":"c3a9799ce44ff4d87e5c2ae1f2181bde6770a205d6e2150766e9d859c880eb61","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:00:13.169104Z","signature_b64":"dKs1/McCf3Rw6VhnHAUxHhdbNkTiUk7grTxXGljWUQgmmfe6OG0bNc79EG5eQiQiw9UvCG8G1K261Noq14tHCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c3a9799ce44ff4d87e5c2ae1f2181bde6770a205d6e2150766e9d859c880eb61","last_reissued_at":"2026-05-18T00:00:13.168474Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:00:13.168474Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1805.08313","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:00:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"51CxR0ES8ubx/cwjN1XJV6wBsBA45SdGEVYhn+wUAmCaYpGMA9m6bVz01bu+kAn7aCHxYzhpr9TpZXNfLVZvCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T16:41:27.466361Z"},"content_sha256":"734858d78a1cb16ca46ba6b9809680bc447372737330256a27f7787886e46d97","schema_version":"1.0","event_id":"sha256:734858d78a1cb16ca46ba6b9809680bc447372737330256a27f7787886e46d97"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:YOUXTHHEJ72NQ7S4FLQ7EGA33Z","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Learning Safe Policies with Expert Guidance","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Doina Precup, Fa Wu, Jessie Huang, Yang Cai","submitted_at":"2018-05-21T22:40:07Z","abstract_excerpt":"We propose a framework for ensuring safe behavior of a reinforcement learning agent when the reward function may be difficult to specify. In order to do this, we rely on the existence of demonstrations from expert policies, and we provide a theoretical framework for the agent to optimize in the space of rewards consistent with its existing knowledge. We propose two methods to solve the resulting optimization: an exact ellipsoid-based method and a method in the spirit of the \"follow-the-perturbed-leader\" algorithm. Our experiments demonstrate the behavior of our algorithm in both discrete and c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1805.08313","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"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":"2026-05-18T00:00:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OMxOLDQLjDgrORxcEtkOK17diRB04wWmoyPKI+2H23TyjsNYhueO/lPuZ6ONDQousI+5FSnLA74sziHmB/q4Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T16:41:27.467056Z"},"content_sha256":"e6116811e889b1bc72f41038c0363b312b11557f3bed0af1e7896eca01aa0658","schema_version":"1.0","event_id":"sha256:e6116811e889b1bc72f41038c0363b312b11557f3bed0af1e7896eca01aa0658"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YOUXTHHEJ72NQ7S4FLQ7EGA33Z/bundle.json","state_url":"https://pith.science/pith/YOUXTHHEJ72NQ7S4FLQ7EGA33Z/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YOUXTHHEJ72NQ7S4FLQ7EGA33Z/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":"2026-05-25T16:41:27Z","links":{"resolver":"https://pith.science/pith/YOUXTHHEJ72NQ7S4FLQ7EGA33Z","bundle":"https://pith.science/pith/YOUXTHHEJ72NQ7S4FLQ7EGA33Z/bundle.json","state":"https://pith.science/pith/YOUXTHHEJ72NQ7S4FLQ7EGA33Z/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YOUXTHHEJ72NQ7S4FLQ7EGA33Z/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:YOUXTHHEJ72NQ7S4FLQ7EGA33Z","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":"1875a140d1ebf49f3c3ee82cd15d02449aff3f750e4fa078cbcdbfb0d71a418a","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-05-21T22:40:07Z","title_canon_sha256":"2e7755388a1c3fbdc166926bb789eeca5ac4dbe03b5b770bfd9fbe9e9d7a6d1d"},"schema_version":"1.0","source":{"id":"1805.08313","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1805.08313","created_at":"2026-05-18T00:00:13Z"},{"alias_kind":"arxiv_version","alias_value":"1805.08313v2","created_at":"2026-05-18T00:00:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1805.08313","created_at":"2026-05-18T00:00:13Z"},{"alias_kind":"pith_short_12","alias_value":"YOUXTHHEJ72N","created_at":"2026-05-18T12:33:04Z"},{"alias_kind":"pith_short_16","alias_value":"YOUXTHHEJ72NQ7S4","created_at":"2026-05-18T12:33:04Z"},{"alias_kind":"pith_short_8","alias_value":"YOUXTHHE","created_at":"2026-05-18T12:33:04Z"}],"graph_snapshots":[{"event_id":"sha256:e6116811e889b1bc72f41038c0363b312b11557f3bed0af1e7896eca01aa0658","target":"graph","created_at":"2026-05-18T00:00:13Z","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"},"paper":{"abstract_excerpt":"We propose a framework for ensuring safe behavior of a reinforcement learning agent when the reward function may be difficult to specify. In order to do this, we rely on the existence of demonstrations from expert policies, and we provide a theoretical framework for the agent to optimize in the space of rewards consistent with its existing knowledge. We propose two methods to solve the resulting optimization: an exact ellipsoid-based method and a method in the spirit of the \"follow-the-perturbed-leader\" algorithm. Our experiments demonstrate the behavior of our algorithm in both discrete and c","authors_text":"Doina Precup, Fa Wu, Jessie Huang, Yang Cai","cross_cats":["cs.AI","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-05-21T22:40:07Z","title":"Learning Safe Policies with Expert Guidance"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1805.08313","kind":"arxiv","version":2},"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":"sha256:734858d78a1cb16ca46ba6b9809680bc447372737330256a27f7787886e46d97","target":"record","created_at":"2026-05-18T00:00:13Z","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":"1875a140d1ebf49f3c3ee82cd15d02449aff3f750e4fa078cbcdbfb0d71a418a","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-05-21T22:40:07Z","title_canon_sha256":"2e7755388a1c3fbdc166926bb789eeca5ac4dbe03b5b770bfd9fbe9e9d7a6d1d"},"schema_version":"1.0","source":{"id":"1805.08313","kind":"arxiv","version":2}},"canonical_sha256":"c3a9799ce44ff4d87e5c2ae1f2181bde6770a205d6e2150766e9d859c880eb61","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c3a9799ce44ff4d87e5c2ae1f2181bde6770a205d6e2150766e9d859c880eb61","first_computed_at":"2026-05-18T00:00:13.168474Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:00:13.168474Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"dKs1/McCf3Rw6VhnHAUxHhdbNkTiUk7grTxXGljWUQgmmfe6OG0bNc79EG5eQiQiw9UvCG8G1K261Noq14tHCg==","signature_status":"signed_v1","signed_at":"2026-05-18T00:00:13.169104Z","signed_message":"canonical_sha256_bytes"},"source_id":"1805.08313","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:734858d78a1cb16ca46ba6b9809680bc447372737330256a27f7787886e46d97","sha256:e6116811e889b1bc72f41038c0363b312b11557f3bed0af1e7896eca01aa0658"],"state_sha256":"e30e4d1c5cf22000c683d85a92cf4c80bd42aaeb58fca5db433e818e6b60f1b5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GRiO4a7wc58NPSZq/CgBNMdy+T2Cq+Kn1/0BC1pFk4rbzyZIAmtTBlv9JvDtoI3c2XO4AlI9X91AJlaCR0KVCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-25T16:41:27.470928Z","bundle_sha256":"3f5bc15006a8a54b9ad9fdd454bf0112ceeee2547c885e9c998c42f8aa5c47b4"}}