{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:33JLWFQHSGLQTTC7J5FGPWPPGV","short_pith_number":"pith:33JLWFQH","canonical_record":{"source":{"id":"1804.02477","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-04-06T22:17:18Z","cross_cats_sorted":["cs.AI","cs.PL","stat.ML"],"title_canon_sha256":"465f9bbb55ad8e1194044ec5d551432cb869c617591ccda0c6fc3d0de8f5d1aa","abstract_canon_sha256":"b91c75b4a42daa045482ea512036f96b71c7893216e600bf8a36fda4d9e45697"},"schema_version":"1.0"},"canonical_sha256":"ded2bb1607919709cc5f4f4a67d9ef354fe3591b047242380834cda1560c8adb","source":{"kind":"arxiv","id":"1804.02477","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1804.02477","created_at":"2026-05-17T23:48:55Z"},{"alias_kind":"arxiv_version","alias_value":"1804.02477v3","created_at":"2026-05-17T23:48:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1804.02477","created_at":"2026-05-17T23:48:55Z"},{"alias_kind":"pith_short_12","alias_value":"33JLWFQHSGLQ","created_at":"2026-05-18T12:32:02Z"},{"alias_kind":"pith_short_16","alias_value":"33JLWFQHSGLQTTC7","created_at":"2026-05-18T12:32:02Z"},{"alias_kind":"pith_short_8","alias_value":"33JLWFQH","created_at":"2026-05-18T12:32:02Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:33JLWFQHSGLQTTC7J5FGPWPPGV","target":"record","payload":{"canonical_record":{"source":{"id":"1804.02477","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-04-06T22:17:18Z","cross_cats_sorted":["cs.AI","cs.PL","stat.ML"],"title_canon_sha256":"465f9bbb55ad8e1194044ec5d551432cb869c617591ccda0c6fc3d0de8f5d1aa","abstract_canon_sha256":"b91c75b4a42daa045482ea512036f96b71c7893216e600bf8a36fda4d9e45697"},"schema_version":"1.0"},"canonical_sha256":"ded2bb1607919709cc5f4f4a67d9ef354fe3591b047242380834cda1560c8adb","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:48:55.615704Z","signature_b64":"3idl0I8cIN5DAV7g8d6mQr7+uYreJ67q4E2/pY29ob8ljgqBeAe504SevRqe0oV/wu8kQPn9zodcZbrQcNvzDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ded2bb1607919709cc5f4f4a67d9ef354fe3591b047242380834cda1560c8adb","last_reissued_at":"2026-05-17T23:48:55.615171Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:48:55.615171Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1804.02477","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":"2026-05-17T23:48:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HJv5j69Pp5hbtOdt4grssK1yAu4v+gFpB4QW7BEH7RX7DISrn95BxrfCwyF25vMIDVb/USr7eu1XhXc3IP+aDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T11:39:03.331879Z"},"content_sha256":"47ff2c8f7ceab869c96d4ce052651897e616a6319b4036eb0f82d4b8e82218b7","schema_version":"1.0","event_id":"sha256:47ff2c8f7ceab869c96d4ce052651897e616a6319b4036eb0f82d4b8e82218b7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:33JLWFQHSGLQTTC7J5FGPWPPGV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Programmatically Interpretable Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.PL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Abhinav Verma, Pushmeet Kohli, Rishabh Singh, Swarat Chaudhuri, Vijayaraghavan Murali","submitted_at":"2018-04-06T22:17:18Z","abstract_excerpt":"We present a reinforcement learning framework, called Programmatically Interpretable Reinforcement Learning (PIRL), that is designed to generate interpretable and verifiable agent policies. Unlike the popular Deep Reinforcement Learning (DRL) paradigm, which represents policies by neural networks, PIRL represents policies using a high-level, domain-specific programming language. Such programmatic policies have the benefits of being more easily interpreted than neural networks, and being amenable to verification by symbolic methods. We propose a new method, called Neurally Directed Program Sear"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1804.02477","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":""},"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":"2026-05-17T23:48:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vBkj+Zmmbun+KrfWuIFsczaDIvJnQkHejOfHFq1cADTKfI/OeoIkwUVKjhMCD/hk5t9Q77dWz/vN74hOujoKCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T11:39:03.332607Z"},"content_sha256":"d4308abd418a53b9e7ecde43426469ff31bb66b7e120563a1e2e1a47185311d6","schema_version":"1.0","event_id":"sha256:d4308abd418a53b9e7ecde43426469ff31bb66b7e120563a1e2e1a47185311d6"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/33JLWFQHSGLQTTC7J5FGPWPPGV/bundle.json","state_url":"https://pith.science/pith/33JLWFQHSGLQTTC7J5FGPWPPGV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/33JLWFQHSGLQTTC7J5FGPWPPGV/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":"2026-05-27T11:39:03Z","links":{"resolver":"https://pith.science/pith/33JLWFQHSGLQTTC7J5FGPWPPGV","bundle":"https://pith.science/pith/33JLWFQHSGLQTTC7J5FGPWPPGV/bundle.json","state":"https://pith.science/pith/33JLWFQHSGLQTTC7J5FGPWPPGV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/33JLWFQHSGLQTTC7J5FGPWPPGV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:33JLWFQHSGLQTTC7J5FGPWPPGV","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":"b91c75b4a42daa045482ea512036f96b71c7893216e600bf8a36fda4d9e45697","cross_cats_sorted":["cs.AI","cs.PL","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-04-06T22:17:18Z","title_canon_sha256":"465f9bbb55ad8e1194044ec5d551432cb869c617591ccda0c6fc3d0de8f5d1aa"},"schema_version":"1.0","source":{"id":"1804.02477","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1804.02477","created_at":"2026-05-17T23:48:55Z"},{"alias_kind":"arxiv_version","alias_value":"1804.02477v3","created_at":"2026-05-17T23:48:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1804.02477","created_at":"2026-05-17T23:48:55Z"},{"alias_kind":"pith_short_12","alias_value":"33JLWFQHSGLQ","created_at":"2026-05-18T12:32:02Z"},{"alias_kind":"pith_short_16","alias_value":"33JLWFQHSGLQTTC7","created_at":"2026-05-18T12:32:02Z"},{"alias_kind":"pith_short_8","alias_value":"33JLWFQH","created_at":"2026-05-18T12:32:02Z"}],"graph_snapshots":[{"event_id":"sha256:d4308abd418a53b9e7ecde43426469ff31bb66b7e120563a1e2e1a47185311d6","target":"graph","created_at":"2026-05-17T23:48:55Z","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 present a reinforcement learning framework, called Programmatically Interpretable Reinforcement Learning (PIRL), that is designed to generate interpretable and verifiable agent policies. Unlike the popular Deep Reinforcement Learning (DRL) paradigm, which represents policies by neural networks, PIRL represents policies using a high-level, domain-specific programming language. Such programmatic policies have the benefits of being more easily interpreted than neural networks, and being amenable to verification by symbolic methods. We propose a new method, called Neurally Directed Program Sear","authors_text":"Abhinav Verma, Pushmeet Kohli, Rishabh Singh, Swarat Chaudhuri, Vijayaraghavan Murali","cross_cats":["cs.AI","cs.PL","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-04-06T22:17:18Z","title":"Programmatically Interpretable Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1804.02477","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":"sha256:47ff2c8f7ceab869c96d4ce052651897e616a6319b4036eb0f82d4b8e82218b7","target":"record","created_at":"2026-05-17T23:48:55Z","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":"b91c75b4a42daa045482ea512036f96b71c7893216e600bf8a36fda4d9e45697","cross_cats_sorted":["cs.AI","cs.PL","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-04-06T22:17:18Z","title_canon_sha256":"465f9bbb55ad8e1194044ec5d551432cb869c617591ccda0c6fc3d0de8f5d1aa"},"schema_version":"1.0","source":{"id":"1804.02477","kind":"arxiv","version":3}},"canonical_sha256":"ded2bb1607919709cc5f4f4a67d9ef354fe3591b047242380834cda1560c8adb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ded2bb1607919709cc5f4f4a67d9ef354fe3591b047242380834cda1560c8adb","first_computed_at":"2026-05-17T23:48:55.615171Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:48:55.615171Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3idl0I8cIN5DAV7g8d6mQr7+uYreJ67q4E2/pY29ob8ljgqBeAe504SevRqe0oV/wu8kQPn9zodcZbrQcNvzDQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:48:55.615704Z","signed_message":"canonical_sha256_bytes"},"source_id":"1804.02477","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:47ff2c8f7ceab869c96d4ce052651897e616a6319b4036eb0f82d4b8e82218b7","sha256:d4308abd418a53b9e7ecde43426469ff31bb66b7e120563a1e2e1a47185311d6"],"state_sha256":"5e26708e116b660a645e6f95bab50c6f6698b653d8206d76d21931ad223cc163"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uW74ZIh+H+boqIvhs7kAKKEuumGVmxO2AlPPTNxHliNWwTUrJ0tDjIr65MTXOiBgvXCXFn1MbkBiDT7n4JdOCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T11:39:03.336546Z","bundle_sha256":"56e7367c1c6af45f2b30b6e09f5fdb72383d411bc334cf7e2c4bd648af19382e"}}