{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:CLRCJLR66G75ZEXEBT3GQQKXE4","short_pith_number":"pith:CLRCJLR6","canonical_record":{"source":{"id":"1912.01217","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2019-12-03T06:44:48Z","cross_cats_sorted":[],"title_canon_sha256":"0ccd928e943847f1ada8737b9612238b9a2e2d862475c58d6c595e68429c6198","abstract_canon_sha256":"6f170a5128e215a774da0829eca285e2dd1d2a670e14495df707f644ee7fdd4a"},"schema_version":"1.0"},"canonical_sha256":"12e224ae3ef1bfdc92e40cf6684157273dbd5b1f2bfa0c2173fcbfa5b32099b4","source":{"kind":"arxiv","id":"1912.01217","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1912.01217","created_at":"2026-07-05T02:18:27Z"},{"alias_kind":"arxiv_version","alias_value":"1912.01217v2","created_at":"2026-07-05T02:18:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1912.01217","created_at":"2026-07-05T02:18:27Z"},{"alias_kind":"pith_short_12","alias_value":"CLRCJLR66G75","created_at":"2026-07-05T02:18:27Z"},{"alias_kind":"pith_short_16","alias_value":"CLRCJLR66G75ZEXE","created_at":"2026-07-05T02:18:27Z"},{"alias_kind":"pith_short_8","alias_value":"CLRCJLR6","created_at":"2026-07-05T02:18:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:CLRCJLR66G75ZEXEBT3GQQKXE4","target":"record","payload":{"canonical_record":{"source":{"id":"1912.01217","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2019-12-03T06:44:48Z","cross_cats_sorted":[],"title_canon_sha256":"0ccd928e943847f1ada8737b9612238b9a2e2d862475c58d6c595e68429c6198","abstract_canon_sha256":"6f170a5128e215a774da0829eca285e2dd1d2a670e14495df707f644ee7fdd4a"},"schema_version":"1.0"},"canonical_sha256":"12e224ae3ef1bfdc92e40cf6684157273dbd5b1f2bfa0c2173fcbfa5b32099b4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:18:27.974441Z","signature_b64":"b77qbyGH9z7+ny0H5WOPqUqnFGq23JpkPMteNT0aGDhand9svzVdxoLXLD4fJgqKt2gIYml8Z3K6l/8GH4xCBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"12e224ae3ef1bfdc92e40cf6684157273dbd5b1f2bfa0c2173fcbfa5b32099b4","last_reissued_at":"2026-07-05T02:18:27.973986Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:18:27.973986Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1912.01217","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":"2026-07-05T02:18:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CQGQatc30rDO+kmEWkW65d6gkZhCBRzSlqwliK1FyEK0eI8QEAWHCj1N+ngH+Pn93+bJb2lm7eZG2HgaPSTIBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-22T08:27:15.537995Z"},"content_sha256":"a54ccaf2cab5519235a8db814d915879b3518140d558f82ea7a4082ae90f30bc","schema_version":"1.0","event_id":"sha256:a54ccaf2cab5519235a8db814d915879b3518140d558f82ea7a4082ae90f30bc"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:CLRCJLR66G75ZEXEBT3GQQKXE4","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SafeLife 1.0: Exploring Side Effects in Complex Environments","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Carroll L. Wainwright, Peter Eckersley","submitted_at":"2019-12-03T06:44:48Z","abstract_excerpt":"We present SafeLife, a publicly available reinforcement learning environment that tests the safety of reinforcement learning agents. It contains complex, dynamic, tunable, procedurally generated levels with many opportunities for unsafe behavior. Agents are graded both on their ability to maximize their explicit reward and on their ability to operate safely without unnecessary side effects. We train agents to maximize rewards using proximal policy optimization and score them on a suite of benchmark levels. The resulting agents are performant but not safe -- they tend to cause large side effect"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1912.01217","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/1912.01217/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":"2026-07-05T02:18:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uHoWbwHZF4L9mLGtD07WxQWQ45UE0OMeQwKObIu2vheuXNoUL+KHOHHK7GU+WKz58Ns42LHVGVBZ62iK0RTkDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-22T08:27:15.538523Z"},"content_sha256":"59ddf3975202e45eb4a55665ec185081687ac4785a9000c143931a8484b2ca92","schema_version":"1.0","event_id":"sha256:59ddf3975202e45eb4a55665ec185081687ac4785a9000c143931a8484b2ca92"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CLRCJLR66G75ZEXEBT3GQQKXE4/bundle.json","state_url":"https://pith.science/pith/CLRCJLR66G75ZEXEBT3GQQKXE4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CLRCJLR66G75ZEXEBT3GQQKXE4/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":"2026-08-22T08:27:15Z","links":{"resolver":"https://pith.science/pith/CLRCJLR66G75ZEXEBT3GQQKXE4","bundle":"https://pith.science/pith/CLRCJLR66G75ZEXEBT3GQQKXE4/bundle.json","state":"https://pith.science/pith/CLRCJLR66G75ZEXEBT3GQQKXE4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CLRCJLR66G75ZEXEBT3GQQKXE4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:CLRCJLR66G75ZEXEBT3GQQKXE4","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":"6f170a5128e215a774da0829eca285e2dd1d2a670e14495df707f644ee7fdd4a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2019-12-03T06:44:48Z","title_canon_sha256":"0ccd928e943847f1ada8737b9612238b9a2e2d862475c58d6c595e68429c6198"},"schema_version":"1.0","source":{"id":"1912.01217","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1912.01217","created_at":"2026-07-05T02:18:27Z"},{"alias_kind":"arxiv_version","alias_value":"1912.01217v2","created_at":"2026-07-05T02:18:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1912.01217","created_at":"2026-07-05T02:18:27Z"},{"alias_kind":"pith_short_12","alias_value":"CLRCJLR66G75","created_at":"2026-07-05T02:18:27Z"},{"alias_kind":"pith_short_16","alias_value":"CLRCJLR66G75ZEXE","created_at":"2026-07-05T02:18:27Z"},{"alias_kind":"pith_short_8","alias_value":"CLRCJLR6","created_at":"2026-07-05T02:18:27Z"}],"graph_snapshots":[{"event_id":"sha256:59ddf3975202e45eb4a55665ec185081687ac4785a9000c143931a8484b2ca92","target":"graph","created_at":"2026-07-05T02:18:27Z","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/1912.01217/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present SafeLife, a publicly available reinforcement learning environment that tests the safety of reinforcement learning agents. It contains complex, dynamic, tunable, procedurally generated levels with many opportunities for unsafe behavior. Agents are graded both on their ability to maximize their explicit reward and on their ability to operate safely without unnecessary side effects. We train agents to maximize rewards using proximal policy optimization and score them on a suite of benchmark levels. The resulting agents are performant but not safe -- they tend to cause large side effect","authors_text":"Carroll L. Wainwright, Peter Eckersley","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2019-12-03T06:44:48Z","title":"SafeLife 1.0: Exploring Side Effects in Complex Environments"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1912.01217","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:a54ccaf2cab5519235a8db814d915879b3518140d558f82ea7a4082ae90f30bc","target":"record","created_at":"2026-07-05T02:18:27Z","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":"6f170a5128e215a774da0829eca285e2dd1d2a670e14495df707f644ee7fdd4a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2019-12-03T06:44:48Z","title_canon_sha256":"0ccd928e943847f1ada8737b9612238b9a2e2d862475c58d6c595e68429c6198"},"schema_version":"1.0","source":{"id":"1912.01217","kind":"arxiv","version":2}},"canonical_sha256":"12e224ae3ef1bfdc92e40cf6684157273dbd5b1f2bfa0c2173fcbfa5b32099b4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"12e224ae3ef1bfdc92e40cf6684157273dbd5b1f2bfa0c2173fcbfa5b32099b4","first_computed_at":"2026-07-05T02:18:27.973986Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:18:27.973986Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"b77qbyGH9z7+ny0H5WOPqUqnFGq23JpkPMteNT0aGDhand9svzVdxoLXLD4fJgqKt2gIYml8Z3K6l/8GH4xCBg==","signature_status":"signed_v1","signed_at":"2026-07-05T02:18:27.974441Z","signed_message":"canonical_sha256_bytes"},"source_id":"1912.01217","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a54ccaf2cab5519235a8db814d915879b3518140d558f82ea7a4082ae90f30bc","sha256:59ddf3975202e45eb4a55665ec185081687ac4785a9000c143931a8484b2ca92"],"state_sha256":"0ac20a2f89cedf0acaace2f2f199ea094125d58fb7d13d4431f50a1e27642e74"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YMcnBjgE5eBNpYY6t64rKHDi56jO+9kKmfhhrxP2BdD67+UvcuKisEO7JqSkHem6+aEpsmeEQgHLgU+A0fhZCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-22T08:27:15.542117Z","bundle_sha256":"e4f51428d4d1e45820c0e806e756008218e48c74ac31830681b5a45670c60b07"}}