{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:6NWEVKWMEH5ASZHZ7MHXNU3MYS","short_pith_number":"pith:6NWEVKWM","canonical_record":{"source":{"id":"2506.22566","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-27T18:28:41Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"37c1ea85ec37718df16a488994d3f0e12b79aa226f6e5c15e7254402e831347e","abstract_canon_sha256":"27f3d8482717a56ae311747e17142a206b593033fc687719c557d5d32685002b"},"schema_version":"1.0"},"canonical_sha256":"f36c4aaacc21fa0964f9fb0f76d36cc48ffd60e10cbbac16c553cd1783a056d2","source":{"kind":"arxiv","id":"2506.22566","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.22566","created_at":"2026-07-05T11:42:54Z"},{"alias_kind":"arxiv_version","alias_value":"2506.22566v3","created_at":"2026-07-05T11:42:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.22566","created_at":"2026-07-05T11:42:54Z"},{"alias_kind":"pith_short_12","alias_value":"6NWEVKWMEH5A","created_at":"2026-07-05T11:42:54Z"},{"alias_kind":"pith_short_16","alias_value":"6NWEVKWMEH5ASZHZ","created_at":"2026-07-05T11:42:54Z"},{"alias_kind":"pith_short_8","alias_value":"6NWEVKWM","created_at":"2026-07-05T11:42:54Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:6NWEVKWMEH5ASZHZ7MHXNU3MYS","target":"record","payload":{"canonical_record":{"source":{"id":"2506.22566","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-27T18:28:41Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"37c1ea85ec37718df16a488994d3f0e12b79aa226f6e5c15e7254402e831347e","abstract_canon_sha256":"27f3d8482717a56ae311747e17142a206b593033fc687719c557d5d32685002b"},"schema_version":"1.0"},"canonical_sha256":"f36c4aaacc21fa0964f9fb0f76d36cc48ffd60e10cbbac16c553cd1783a056d2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:42:54.860805Z","signature_b64":"nVGQERBO0HGaD213iNjgW5pqm+PF2UvBKmybijryUUE5vjxKQ6aA+srkjZGoddz1mhNDJGHtx2nap2XrcxY8DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f36c4aaacc21fa0964f9fb0f76d36cc48ffd60e10cbbac16c553cd1783a056d2","last_reissued_at":"2026-07-05T11:42:54.860288Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:42:54.860288Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.22566","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":"2026-07-05T11:42:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3MkCo8oWgYKUQS7IUXv0eGHCjJ8X1Q01ioY0v8ZWmUh33nw1Jg9FBXD/wnbsyVgXy/qbdy7OvLi4WGWGRwkNCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T03:22:16.748159Z"},"content_sha256":"98f523d2f5105ea3e325d6ee19cb1e5b05873c530889d17e2cd18271279f432d","schema_version":"1.0","event_id":"sha256:98f523d2f5105ea3e325d6ee19cb1e5b05873c530889d17e2cd18271279f432d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:6NWEVKWMEH5ASZHZ7MHXNU3MYS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Exploration Behavior of Untrained Policies","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Jacob Adamczyk","submitted_at":"2025-06-27T18:28:41Z","abstract_excerpt":"Exploration remains a fundamental challenge in reinforcement learning (RL), particularly in environments with sparse or adversarial reward structures. In this work, we study how the architecture of deep neural policies implicitly shapes exploration before training. We theoretically and empirically demonstrate strategies for generating ballistic or diffusive trajectories from untrained policies in a toy model. Using the theory of infinite-width networks and a continuous-time limit, we show that untrained policies return correlated actions and result in non-trivial state-visitation distributions"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.22566","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2506.22566/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":"2026-07-05T11:42:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"we4yROBxt45nsfUvf/PtVxmIZvA50N91UnLnkBkqFeFqHKiJAG8A0N/uFJimqOyDIcr4htJtcVvtwfhoO+RmAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T03:22:16.748788Z"},"content_sha256":"98f0ef5196f100733625d416ca02e5514078dc19dc597815439ce39c67981192","schema_version":"1.0","event_id":"sha256:98f0ef5196f100733625d416ca02e5514078dc19dc597815439ce39c67981192"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6NWEVKWMEH5ASZHZ7MHXNU3MYS/bundle.json","state_url":"https://pith.science/pith/6NWEVKWMEH5ASZHZ7MHXNU3MYS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6NWEVKWMEH5ASZHZ7MHXNU3MYS/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":"2026-08-11T03:22:16Z","links":{"resolver":"https://pith.science/pith/6NWEVKWMEH5ASZHZ7MHXNU3MYS","bundle":"https://pith.science/pith/6NWEVKWMEH5ASZHZ7MHXNU3MYS/bundle.json","state":"https://pith.science/pith/6NWEVKWMEH5ASZHZ7MHXNU3MYS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6NWEVKWMEH5ASZHZ7MHXNU3MYS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:6NWEVKWMEH5ASZHZ7MHXNU3MYS","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":"27f3d8482717a56ae311747e17142a206b593033fc687719c557d5d32685002b","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-27T18:28:41Z","title_canon_sha256":"37c1ea85ec37718df16a488994d3f0e12b79aa226f6e5c15e7254402e831347e"},"schema_version":"1.0","source":{"id":"2506.22566","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.22566","created_at":"2026-07-05T11:42:54Z"},{"alias_kind":"arxiv_version","alias_value":"2506.22566v3","created_at":"2026-07-05T11:42:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.22566","created_at":"2026-07-05T11:42:54Z"},{"alias_kind":"pith_short_12","alias_value":"6NWEVKWMEH5A","created_at":"2026-07-05T11:42:54Z"},{"alias_kind":"pith_short_16","alias_value":"6NWEVKWMEH5ASZHZ","created_at":"2026-07-05T11:42:54Z"},{"alias_kind":"pith_short_8","alias_value":"6NWEVKWM","created_at":"2026-07-05T11:42:54Z"}],"graph_snapshots":[{"event_id":"sha256:98f0ef5196f100733625d416ca02e5514078dc19dc597815439ce39c67981192","target":"graph","created_at":"2026-07-05T11:42:54Z","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/2506.22566/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Exploration remains a fundamental challenge in reinforcement learning (RL), particularly in environments with sparse or adversarial reward structures. In this work, we study how the architecture of deep neural policies implicitly shapes exploration before training. We theoretically and empirically demonstrate strategies for generating ballistic or diffusive trajectories from untrained policies in a toy model. Using the theory of infinite-width networks and a continuous-time limit, we show that untrained policies return correlated actions and result in non-trivial state-visitation distributions","authors_text":"Jacob Adamczyk","cross_cats":["cs.AI","stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-27T18:28:41Z","title":"Exploration Behavior of Untrained Policies"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.22566","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":"sha256:98f523d2f5105ea3e325d6ee19cb1e5b05873c530889d17e2cd18271279f432d","target":"record","created_at":"2026-07-05T11:42:54Z","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":"27f3d8482717a56ae311747e17142a206b593033fc687719c557d5d32685002b","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-27T18:28:41Z","title_canon_sha256":"37c1ea85ec37718df16a488994d3f0e12b79aa226f6e5c15e7254402e831347e"},"schema_version":"1.0","source":{"id":"2506.22566","kind":"arxiv","version":3}},"canonical_sha256":"f36c4aaacc21fa0964f9fb0f76d36cc48ffd60e10cbbac16c553cd1783a056d2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f36c4aaacc21fa0964f9fb0f76d36cc48ffd60e10cbbac16c553cd1783a056d2","first_computed_at":"2026-07-05T11:42:54.860288Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:42:54.860288Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"nVGQERBO0HGaD213iNjgW5pqm+PF2UvBKmybijryUUE5vjxKQ6aA+srkjZGoddz1mhNDJGHtx2nap2XrcxY8DQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:42:54.860805Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.22566","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:98f523d2f5105ea3e325d6ee19cb1e5b05873c530889d17e2cd18271279f432d","sha256:98f0ef5196f100733625d416ca02e5514078dc19dc597815439ce39c67981192"],"state_sha256":"4b201004b84ac0d61afe52fc739ad7682a29df3b9c58703a59aef0d872d64bdc"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3LoLWtCJFjSxpRLRa/3tp3cJDr3ze058tq7/QxzCdY09Eqj7woMFHbXs21ShcEKwLumq4L3VfwaIQuORcRScCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T03:22:16.825940Z","bundle_sha256":"aed012f138b3fac4ae4f2af729329085cc1fb6be4e5cf7d7c39bf0cea49da6f9"}}