{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:NYG37TM4FQCD2Z3DYUB5UWPCG4","short_pith_number":"pith:NYG37TM4","schema_version":"1.0","canonical_sha256":"6e0dbfcd9c2c043d6763c503da59e2370dcb8caf9235b490a7a96512874cca92","source":{"kind":"arxiv","id":"2309.15676","version":1},"attestation_state":"computed","paper":{"title":"Joint Sampling and Optimisation for Inverse Rendering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.GR","authors_text":"Gurprit Singh, Hans-Peter Seidel, Karol Myszkowski, Martin Balint","submitted_at":"2023-09-27T14:21:13Z","abstract_excerpt":"When dealing with difficult inverse problems such as inverse rendering, using Monte Carlo estimated gradients to optimise parameters can slow down convergence due to variance. Averaging many gradient samples in each iteration reduces this variance trivially. However, for problems that require thousands of optimisation iterations, the computational cost of this approach rises quickly.\n  We derive a theoretical framework for interleaving sampling and optimisation. We update and reuse past samples with low-variance finite-difference estimators that describe the change in the estimated gradients b"},"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":"2309.15676","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.GR","submitted_at":"2023-09-27T14:21:13Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"c0db2b20b5855c3675d3f23bc092b16d541bc3083a9d7f38cbde99cdd17229ed","abstract_canon_sha256":"66097447278da5389d3b816ab4873c3820b3f034634c9c7c44b9bf2520226032"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:54:56.377378Z","signature_b64":"oReuH3UQVRpA+rreUG5icWJkcrYApuIXopmYLBy8QNe/4IUehk9wYw5Oj5EuYlMCcEqR5KzBCFDdmKLmPe7yAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6e0dbfcd9c2c043d6763c503da59e2370dcb8caf9235b490a7a96512874cca92","last_reissued_at":"2026-07-05T06:54:56.376891Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:54:56.376891Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Joint Sampling and Optimisation for Inverse Rendering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.GR","authors_text":"Gurprit Singh, Hans-Peter Seidel, Karol Myszkowski, Martin Balint","submitted_at":"2023-09-27T14:21:13Z","abstract_excerpt":"When dealing with difficult inverse problems such as inverse rendering, using Monte Carlo estimated gradients to optimise parameters can slow down convergence due to variance. Averaging many gradient samples in each iteration reduces this variance trivially. However, for problems that require thousands of optimisation iterations, the computational cost of this approach rises quickly.\n  We derive a theoretical framework for interleaving sampling and optimisation. We update and reuse past samples with low-variance finite-difference estimators that describe the change in the estimated gradients b"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.15676","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/2309.15676/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":"2309.15676","created_at":"2026-07-05T06:54:56.376951+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.15676v1","created_at":"2026-07-05T06:54:56.376951+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.15676","created_at":"2026-07-05T06:54:56.376951+00:00"},{"alias_kind":"pith_short_12","alias_value":"NYG37TM4FQCD","created_at":"2026-07-05T06:54:56.376951+00:00"},{"alias_kind":"pith_short_16","alias_value":"NYG37TM4FQCD2Z3D","created_at":"2026-07-05T06:54:56.376951+00:00"},{"alias_kind":"pith_short_8","alias_value":"NYG37TM4","created_at":"2026-07-05T06:54:56.376951+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/NYG37TM4FQCD2Z3DYUB5UWPCG4","json":"https://pith.science/pith/NYG37TM4FQCD2Z3DYUB5UWPCG4.json","graph_json":"https://pith.science/api/pith-number/NYG37TM4FQCD2Z3DYUB5UWPCG4/graph.json","events_json":"https://pith.science/api/pith-number/NYG37TM4FQCD2Z3DYUB5UWPCG4/events.json","paper":"https://pith.science/paper/NYG37TM4"},"agent_actions":{"view_html":"https://pith.science/pith/NYG37TM4FQCD2Z3DYUB5UWPCG4","download_json":"https://pith.science/pith/NYG37TM4FQCD2Z3DYUB5UWPCG4.json","view_paper":"https://pith.science/paper/NYG37TM4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.15676&json=true","fetch_graph":"https://pith.science/api/pith-number/NYG37TM4FQCD2Z3DYUB5UWPCG4/graph.json","fetch_events":"https://pith.science/api/pith-number/NYG37TM4FQCD2Z3DYUB5UWPCG4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NYG37TM4FQCD2Z3DYUB5UWPCG4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NYG37TM4FQCD2Z3DYUB5UWPCG4/action/storage_attestation","attest_author":"https://pith.science/pith/NYG37TM4FQCD2Z3DYUB5UWPCG4/action/author_attestation","sign_citation":"https://pith.science/pith/NYG37TM4FQCD2Z3DYUB5UWPCG4/action/citation_signature","submit_replication":"https://pith.science/pith/NYG37TM4FQCD2Z3DYUB5UWPCG4/action/replication_record"}},"created_at":"2026-07-05T06:54:56.376951+00:00","updated_at":"2026-07-05T06:54:56.376951+00:00"}