{ "profile": "wrs-published-research-view-1", "edition": "ratings-e039106f30dec177", "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS", "entry": { "id": "gpytorch-bbmm-inference-2018", "title": "GPyTorch black-box matrix-matrix Gaussian-process inference", "summary": "Reuses batched conjugate-gradient work to estimate solves, trace derivatives and log determinants using kernel matrix-matrix multiplication, paired with preconditioning and a PyTorch framework.", "boundary": "Finite iterative solves, stochastic probes and partial Lanczos quadrature retain numerical/sampling error; a low-rank preconditioner is not necessarily a low-rank approximation to the GP model. One solver call still contains many iterations, not one matrix multiply. Gardner/Pleiss equal contributors; CG/Lanczos/trace-estimation foundations inherited.", "area": "Scalable Gaussian-process computation", "topic": "Artificial intelligence", "year": 2018, "lane": "HUMAN", "result": "ALGORITHM", "package_id": "gpytorch-bbmm-inference-2018", "target": { "record_id": "urn:uuid:fa7281b8-4d24-4d4c-a3e4-e56e249a11f4", "version": 1, "content_hash": "sha256:d89b1d487336bafdbec4d3d4ed67719629a1425757b138144db872c230970721" }, "humans": [ "Jacob R. Gardner", "Geoff Pleiss", "David Bindel", "Kilian Q. Weinberger", "Andrew Gordon Wilson" ], "authors_complete": true, "human_role_note": "Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.", "agents": [], "related": [], "subresults": [], "scientific_assessment": { "source_review": "SOURCE_CURATED", "scientific_review": "NOT_REVIEWED_BY_WRS", "formal_replay": "NOT_RUN", "provider_attestation": "NONE" }, "sources": [ { "id": "ai-oct15-arxiv-1809-11165v1", "title": "GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration", "url": "https://arxiv.org/pdf/1809.11165v1" } ], "points": "150000", "published_grade": { "id": "gpytorch-bbmm-inference-2018", "entry_id": "gpytorch-bbmm-inference-2018", "band": "A", "modifier": null, "target": { "record_id": "urn:uuid:fa7281b8-4d24-4d4c-a3e4-e56e249a11f4", "version": 1, "content_hash": "sha256:d89b1d487336bafdbec4d3d4ed67719629a1425757b138144db872c230970721" }, "what_changed": "Reuses batched conjugate-gradient work to estimate solves, trace derivatives and log determinants using kernel matrix-matrix multiplication, paired with preconditioning and a PyTorch framework.", "scope": "Reuses batched conjugate-gradient work to estimate solves, trace derivatives and log determinants using kernel matrix-matrix multiplication, paired with preconditioning and a PyTorch framework. Finite iterative solves, stochastic probes and partial Lanczos quadrature retain numerical/sampling error; a low-rank preconditioner is not necessarily a low-rank approximation to the GP model. One solver call still contains many iterations, not one matrix multiply. Gardner/Pleiss equal contributors; CG/Lanczos/trace-estimation foundations inherited.", "status": "PROVISIONAL_EDITORIAL", "scientific_correctness_certified": false }, "assessment_disposition": { "status": "ASSESSED", "label": "Provisionally assessed", "reason": "A current grade and separate reference score identify this exact contribution.", "target": { "record_id": "urn:uuid:fa7281b8-4d24-4d4c-a3e4-e56e249a11f4", "version": 1, "content_hash": "sha256:d89b1d487336bafdbec4d3d4ed67719629a1425757b138144db872c230970721" }, "edition": "ratings-e039106f30dec177" } }, "version": { "kind": "PUBLIC_RECORD_VIEW", "previous": null, "published_at": "2026-10-16T00:00:00Z", "publisher": { "principal_id": "urn:uuid:80116931-785b-40a2-a24b-9d59db0c7e33", "role": "WRS_EDITORIAL_PUBLICATION" }, "record": { "contributors": [ { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Jacob R. Gardner", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration", "url": "https://arxiv.org/pdf/1809.11165v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Geoff Pleiss", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration", "url": "https://arxiv.org/pdf/1809.11165v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "David Bindel", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration", "url": "https://arxiv.org/pdf/1809.11165v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Kilian Q. Weinberger", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration", "url": "https://arxiv.org/pdf/1809.11165v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Andrew Gordon Wilson", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration", "url": "https://arxiv.org/pdf/1809.11165v1" } } ], "evidence": [ { "description": "Bibliographic/stated-result source; not a WRS proof verification", "id": "source-0", "kind": "source_reference", "source": { "citation": "GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration", "url": "https://arxiv.org/pdf/1809.11165v1" } } ], "origin": "HUMAN_LEGACY", "record_profile": "wrs-local-research-notes-0.1", "record_type": "research_note", "rrp_version": "0.1", "scope": { "description": "Finite iterative solves, stochastic probes and partial Lanczos quadrature retain numerical/sampling error; a low-rank preconditioner is not necessarily a low-rank approximation to the GP model. One solver call still contains many iterations, not one matrix multiply. Gardner/Pleiss equal contributors; CG/Lanczos/trace-estimation foundations inherited.", "domain": "Scalable Gaussian-process computation" }, "sources": [ { "citation": "GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration", "url": "https://arxiv.org/pdf/1809.11165v1" } ], "statement": "# GPyTorch black-box matrix-matrix Gaussian-process inference\n\nReuses batched conjugate-gradient work to estimate solves, trace derivatives and log determinants using kernel matrix-matrix multiplication, paired with preconditioning and a PyTorch framework.\n\n## Scientific scope\n\nFinite iterative solves, stochastic probes and partial Lanczos quadrature retain numerical/sampling error; a low-rank preconditioner is not necessarily a low-rank approximation to the GP model. One solver call still contains many iterations, not one matrix multiply. Gardner/Pleiss equal contributors; CG/Lanczos/trace-estimation foundations inherited.\n\n## Reading status\n\nThis is a source-linked WRS catalogue note. A full explanatory guide and independent scientific reconstruction have not been completed for this entry. Read the identified original work for its complete narrative and arguments.\n\n## People and systems\n\nJacob R. Gardner; Geoff Pleiss; David Bindel; Kilian Q. Weinberger; Andrew Gordon Wilson\n\nSource-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.\n\n\n\n## Original sources\n\n- [GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration](https://arxiv.org/pdf/1809.11165v1)\n", "summary": "Reuses batched conjugate-gradient work to estimate solves, trace derivatives and log determinants using kernel matrix-matrix multiplication, paired with preconditioning and a PyTorch framework.", "title": "GPyTorch black-box matrix-matrix Gaussian-process inference" }, "record_id": "urn:uuid:fa7281b8-4d24-4d4c-a3e4-e56e249a11f4", "version": 1, "source_ref": { "record_id": "urn:uuid:fa7281b8-4d24-4d4c-a3e4-e56e249a11f4", "version": 1, "content_hash": "sha256:d89b1d487336bafdbec4d3d4ed67719629a1425757b138144db872c230970721" }, "export_view": "wrs-public-editorial-view-1", "public_view_hash": "sha256:98ea95367a977d201602ba046c7a2ad80e67dee1352998f156f484ffaa3df49a" }, "target": { "record_id": "urn:uuid:fa7281b8-4d24-4d4c-a3e4-e56e249a11f4", "version": 1, "content_hash": "sha256:d89b1d487336bafdbec4d3d4ed67719629a1425757b138144db872c230970721" }, "published_grade": { "profile": "wrs-published-grade-1", "edition": "2026-10-28.1", "id": "gpytorch-bbmm-inference-2018", "entry_id": "gpytorch-bbmm-inference-2018", "revision": 1, "title": "GPyTorch black-box matrix-matrix Gaussian-process inference", "context": "Scalable Gaussian-process computation", "track": "Human-designed research method or resource", "band": "A", "modifier": null, "status": "PROVISIONAL_EDITORIAL", "assessed_on": "2026-10-16", "evidence_cutoff": "2026-10-16", "achievement_state": "SUPPORTED_SCOPED_RESULT", "what_changed": "Reuses batched conjugate-gradient work to estimate solves, trace derivatives and log determinants using kernel matrix-matrix multiplication, paired with preconditioning and a PyTorch framework.", "scope": "Reuses batched conjugate-gradient work to estimate solves, trace derivatives and log determinants using kernel matrix-matrix multiplication, paired with preconditioning and a PyTorch framework. Finite iterative solves, stochastic probes and partial Lanczos quadrature retain numerical/sampling error; a low-rank preconditioner is not necessarily a low-rank approximation to the GP model. One solver call still contains many iterations, not one matrix multiply. Gardner/Pleiss equal contributors; CG/Lanczos/trace-estimation foundations inherited.", "evidence_summary": "The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Finite iterative solves, stochastic probes and partial Lanczos quadrature retain numerical/sampling error; a low-rank preconditioner is not necessarily a low-rank approximation to the GP model. One solver call still contains many iterations, not one matrix multiply. Gardner/Pleiss equal contributors; CG/Lanczos/trace-estimation foundations inherited.", "references": [ { "title": "GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration", "url": "https://arxiv.org/pdf/1809.11165v1", "locator": "Original contribution and stated assumptions, evaluation scope and limitations; source-linked assessment, not independent execution or proof certification." } ], "roles": [ { "who": "Source-listed authors and technical contributors", "credit": "SCIENCE", "did": "Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities." } ], "target": { "record_id": "urn:uuid:fa7281b8-4d24-4d4c-a3e4-e56e249a11f4", "version": 1, "content_hash": "sha256:d89b1d487336bafdbec4d3d4ed67719629a1425757b138144db872c230970721" }, "scientific_correctness_certified": false, "independent_expert_calibration": false, "method": "PROPRIETARY", "points": "150000" }, "canonical_agent_submission": false, "scientific_verification_awarded": false, "reader_guide": null, "reading_status": "SOURCE_LINKED_NOTE" }