{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:CZ2PDPLEUZ5ZUYCW3DWI7B352Y","short_pith_number":"pith:CZ2PDPLE","canonical_record":{"source":{"id":"1905.01240","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-05-03T15:59:25Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"68c2d1bdffb073e8fd40e578c7d4775b66baa8a6670c73ce00b81a2df590f24f","abstract_canon_sha256":"7d8cb1af03643bff336f6c131bb2bb4bcac0349624f83242f09dca6bc6f6d409"},"schema_version":"1.0"},"canonical_sha256":"1674f1bd64a67b9a6056d8ec8f877dd6163fdb2e978a64d813a336b102b9a443","source":{"kind":"arxiv","id":"1905.01240","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1905.01240","created_at":"2026-05-17T23:47:06Z"},{"alias_kind":"arxiv_version","alias_value":"1905.01240v1","created_at":"2026-05-17T23:47:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1905.01240","created_at":"2026-05-17T23:47:06Z"},{"alias_kind":"pith_short_12","alias_value":"CZ2PDPLEUZ5Z","created_at":"2026-05-18T12:33:15Z"},{"alias_kind":"pith_short_16","alias_value":"CZ2PDPLEUZ5ZUYCW","created_at":"2026-05-18T12:33:15Z"},{"alias_kind":"pith_short_8","alias_value":"CZ2PDPLE","created_at":"2026-05-18T12:33:15Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:CZ2PDPLEUZ5ZUYCW3DWI7B352Y","target":"record","payload":{"canonical_record":{"source":{"id":"1905.01240","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-05-03T15:59:25Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"68c2d1bdffb073e8fd40e578c7d4775b66baa8a6670c73ce00b81a2df590f24f","abstract_canon_sha256":"7d8cb1af03643bff336f6c131bb2bb4bcac0349624f83242f09dca6bc6f6d409"},"schema_version":"1.0"},"canonical_sha256":"1674f1bd64a67b9a6056d8ec8f877dd6163fdb2e978a64d813a336b102b9a443","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:47:06.436193Z","signature_b64":"x5imzpk9HRSKs5Bgb/h7HXJzgfq8Z6WLPOPFcAw4C5NAVAJ5rJvjUvMEUZxanS/hYzu22m9KL+29O/WkcqmPCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1674f1bd64a67b9a6056d8ec8f877dd6163fdb2e978a64d813a336b102b9a443","last_reissued_at":"2026-05-17T23:47:06.435432Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:47:06.435432Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1905.01240","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":"2026-05-17T23:47:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6x4pKKUIPHmukGGWPnaekXhPyigwKBHe9U/JSdTKwVJabsrIiY0+Toz9Y1WZ9vspBuFI73VrK6z9IDZnVExkBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-28T17:35:23.948391Z"},"content_sha256":"7bdfbcb2dfc178e7b5d066418a9718fe994cf756cd363209d1501cdec4e4fbaf","schema_version":"1.0","event_id":"sha256:7bdfbcb2dfc178e7b5d066418a9718fe994cf756cd363209d1501cdec4e4fbaf"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:CZ2PDPLEUZ5ZUYCW3DWI7B352Y","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Information asymmetry in KL-regularized RL","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexandre Galashov, Dhruva Tirumala, Guillaume Desjardins, Jonathan Schwarz, Leonard Hasenclever, Nicolas Heess, Razvan Pascanu, Siddhant M. Jayakumar, Wojciech M. Czarnecki, Yee Whye Teh","submitted_at":"2019-05-03T15:59:25Z","abstract_excerpt":"Many real world tasks exhibit rich structure that is repeated across different parts of the state space or in time. In this work we study the possibility of leveraging such repeated structure to speed up and regularize learning. We start from the KL regularized expected reward objective which introduces an additional component, a default policy. Instead of relying on a fixed default policy, we learn it from data. But crucially, we restrict the amount of information the default policy receives, forcing it to learn reusable behaviors that help the policy learn faster. We formalize this strategy "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1905.01240","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":""},"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":"2026-05-17T23:47:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LLAb4EDmgZsayNDyd+5AOzzHda0mD3YVXANJUugz1MrFZ5j7D2On3xT+a3HvIhSsSP+Qljuc/3dapjM223grAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-28T17:35:23.949078Z"},"content_sha256":"bc223115725c34bbba296640382cc32d284361661d1ddabde21fcda48f620833","schema_version":"1.0","event_id":"sha256:bc223115725c34bbba296640382cc32d284361661d1ddabde21fcda48f620833"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CZ2PDPLEUZ5ZUYCW3DWI7B352Y/bundle.json","state_url":"https://pith.science/pith/CZ2PDPLEUZ5ZUYCW3DWI7B352Y/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CZ2PDPLEUZ5ZUYCW3DWI7B352Y/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":"2026-05-28T17:35:23Z","links":{"resolver":"https://pith.science/pith/CZ2PDPLEUZ5ZUYCW3DWI7B352Y","bundle":"https://pith.science/pith/CZ2PDPLEUZ5ZUYCW3DWI7B352Y/bundle.json","state":"https://pith.science/pith/CZ2PDPLEUZ5ZUYCW3DWI7B352Y/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CZ2PDPLEUZ5ZUYCW3DWI7B352Y/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:CZ2PDPLEUZ5ZUYCW3DWI7B352Y","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":"7d8cb1af03643bff336f6c131bb2bb4bcac0349624f83242f09dca6bc6f6d409","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-05-03T15:59:25Z","title_canon_sha256":"68c2d1bdffb073e8fd40e578c7d4775b66baa8a6670c73ce00b81a2df590f24f"},"schema_version":"1.0","source":{"id":"1905.01240","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1905.01240","created_at":"2026-05-17T23:47:06Z"},{"alias_kind":"arxiv_version","alias_value":"1905.01240v1","created_at":"2026-05-17T23:47:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1905.01240","created_at":"2026-05-17T23:47:06Z"},{"alias_kind":"pith_short_12","alias_value":"CZ2PDPLEUZ5Z","created_at":"2026-05-18T12:33:15Z"},{"alias_kind":"pith_short_16","alias_value":"CZ2PDPLEUZ5ZUYCW","created_at":"2026-05-18T12:33:15Z"},{"alias_kind":"pith_short_8","alias_value":"CZ2PDPLE","created_at":"2026-05-18T12:33:15Z"}],"graph_snapshots":[{"event_id":"sha256:bc223115725c34bbba296640382cc32d284361661d1ddabde21fcda48f620833","target":"graph","created_at":"2026-05-17T23:47:06Z","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":"Many real world tasks exhibit rich structure that is repeated across different parts of the state space or in time. In this work we study the possibility of leveraging such repeated structure to speed up and regularize learning. We start from the KL regularized expected reward objective which introduces an additional component, a default policy. Instead of relying on a fixed default policy, we learn it from data. But crucially, we restrict the amount of information the default policy receives, forcing it to learn reusable behaviors that help the policy learn faster. We formalize this strategy ","authors_text":"Alexandre Galashov, Dhruva Tirumala, Guillaume Desjardins, Jonathan Schwarz, Leonard Hasenclever, Nicolas Heess, Razvan Pascanu, Siddhant M. Jayakumar, Wojciech M. Czarnecki, Yee Whye Teh","cross_cats":["cs.AI","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-05-03T15:59:25Z","title":"Information asymmetry in KL-regularized RL"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1905.01240","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":"sha256:7bdfbcb2dfc178e7b5d066418a9718fe994cf756cd363209d1501cdec4e4fbaf","target":"record","created_at":"2026-05-17T23:47:06Z","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":"7d8cb1af03643bff336f6c131bb2bb4bcac0349624f83242f09dca6bc6f6d409","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-05-03T15:59:25Z","title_canon_sha256":"68c2d1bdffb073e8fd40e578c7d4775b66baa8a6670c73ce00b81a2df590f24f"},"schema_version":"1.0","source":{"id":"1905.01240","kind":"arxiv","version":1}},"canonical_sha256":"1674f1bd64a67b9a6056d8ec8f877dd6163fdb2e978a64d813a336b102b9a443","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1674f1bd64a67b9a6056d8ec8f877dd6163fdb2e978a64d813a336b102b9a443","first_computed_at":"2026-05-17T23:47:06.435432Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:47:06.435432Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"x5imzpk9HRSKs5Bgb/h7HXJzgfq8Z6WLPOPFcAw4C5NAVAJ5rJvjUvMEUZxanS/hYzu22m9KL+29O/WkcqmPCQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:47:06.436193Z","signed_message":"canonical_sha256_bytes"},"source_id":"1905.01240","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7bdfbcb2dfc178e7b5d066418a9718fe994cf756cd363209d1501cdec4e4fbaf","sha256:bc223115725c34bbba296640382cc32d284361661d1ddabde21fcda48f620833"],"state_sha256":"144a56519fcab8313aafec57fb4bb7865d6cd28b759d0df3fdd5ca0bb6959a72"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dh1p8nU1BnNM/HinSl/NCwf3uLZNGIJZ7Z96rhBjsyUXSs2CzEMl5+EI+Jth3Yai7Hskruz3+7vKXVSy8wOiAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-28T17:35:23.952800Z","bundle_sha256":"27391e5b0f760d1ceeb27af43c3100bd2914a185aaf401c94162417789082956"}}