{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:MYJCOOYF32K45B2B3VFPF7UFSH","short_pith_number":"pith:MYJCOOYF","schema_version":"1.0","canonical_sha256":"6612273b05de95ce8741dd4af2fe8591f11c242f40fa70cdaef9f8ea1bf14f63","source":{"kind":"arxiv","id":"2201.05830","version":1},"attestation_state":"computed","paper":{"title":"Physical Derivatives: Computing policy gradients by physical forward-propagation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.DS","stat.ML"],"primary_cat":"cs.RO","authors_text":"Arash Mehrjou, Ashkan Soleymani, Bernhard Sch\\\"olkopf, Stefan Bauer","submitted_at":"2022-01-15T11:27:42Z","abstract_excerpt":"Model-free and model-based reinforcement learning are two ends of a spectrum. Learning a good policy without a dynamic model can be prohibitively expensive. Learning the dynamic model of a system can reduce the cost of learning the policy, but it can also introduce bias if it is not accurate. We propose a middle ground where instead of the transition model, the sensitivity of the trajectories with respect to the perturbation of the parameters is learned. This allows us to predict the local behavior of the physical system around a set of nominal policies without knowing the actual model. We ass"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2201.05830","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2022-01-15T11:27:42Z","cross_cats_sorted":["math.DS","stat.ML"],"title_canon_sha256":"5716f4ce7c344ed38c65c6e1c4089130a4d463098b1725736b9aa942598c9e78","abstract_canon_sha256":"c438fa4dfa3151cf7ff0149a6a76a16f89ad0852f749ee1b395bc1e24cb3940b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:48:52.083477Z","signature_b64":"wBOOk9X9PUkwj3QboXVu27qM7qA6dG9eKJ4QG63HrNLLrduVdWcEeWHgCSxHs3t9sq+7oH5jsOZd9kaouMabDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6612273b05de95ce8741dd4af2fe8591f11c242f40fa70cdaef9f8ea1bf14f63","last_reissued_at":"2026-07-05T03:48:52.083118Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:48:52.083118Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Physical Derivatives: Computing policy gradients by physical forward-propagation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.DS","stat.ML"],"primary_cat":"cs.RO","authors_text":"Arash Mehrjou, Ashkan Soleymani, Bernhard Sch\\\"olkopf, Stefan Bauer","submitted_at":"2022-01-15T11:27:42Z","abstract_excerpt":"Model-free and model-based reinforcement learning are two ends of a spectrum. Learning a good policy without a dynamic model can be prohibitively expensive. Learning the dynamic model of a system can reduce the cost of learning the policy, but it can also introduce bias if it is not accurate. We propose a middle ground where instead of the transition model, the sensitivity of the trajectories with respect to the perturbation of the parameters is learned. This allows us to predict the local behavior of the physical system around a set of nominal policies without knowing the actual model. We ass"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.05830","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2201.05830/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2201.05830","created_at":"2026-07-05T03:48:52.083168+00:00"},{"alias_kind":"arxiv_version","alias_value":"2201.05830v1","created_at":"2026-07-05T03:48:52.083168+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.05830","created_at":"2026-07-05T03:48:52.083168+00:00"},{"alias_kind":"pith_short_12","alias_value":"MYJCOOYF32K4","created_at":"2026-07-05T03:48:52.083168+00:00"},{"alias_kind":"pith_short_16","alias_value":"MYJCOOYF32K45B2B","created_at":"2026-07-05T03:48:52.083168+00:00"},{"alias_kind":"pith_short_8","alias_value":"MYJCOOYF","created_at":"2026-07-05T03:48:52.083168+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MYJCOOYF32K45B2B3VFPF7UFSH","json":"https://pith.science/pith/MYJCOOYF32K45B2B3VFPF7UFSH.json","graph_json":"https://pith.science/api/pith-number/MYJCOOYF32K45B2B3VFPF7UFSH/graph.json","events_json":"https://pith.science/api/pith-number/MYJCOOYF32K45B2B3VFPF7UFSH/events.json","paper":"https://pith.science/paper/MYJCOOYF"},"agent_actions":{"view_html":"https://pith.science/pith/MYJCOOYF32K45B2B3VFPF7UFSH","download_json":"https://pith.science/pith/MYJCOOYF32K45B2B3VFPF7UFSH.json","view_paper":"https://pith.science/paper/MYJCOOYF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2201.05830&json=true","fetch_graph":"https://pith.science/api/pith-number/MYJCOOYF32K45B2B3VFPF7UFSH/graph.json","fetch_events":"https://pith.science/api/pith-number/MYJCOOYF32K45B2B3VFPF7UFSH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MYJCOOYF32K45B2B3VFPF7UFSH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MYJCOOYF32K45B2B3VFPF7UFSH/action/storage_attestation","attest_author":"https://pith.science/pith/MYJCOOYF32K45B2B3VFPF7UFSH/action/author_attestation","sign_citation":"https://pith.science/pith/MYJCOOYF32K45B2B3VFPF7UFSH/action/citation_signature","submit_replication":"https://pith.science/pith/MYJCOOYF32K45B2B3VFPF7UFSH/action/replication_record"}},"created_at":"2026-07-05T03:48:52.083168+00:00","updated_at":"2026-07-05T03:48:52.083168+00:00"}