{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:75362SEHMIURXS4OSZ6UAZLLHH","short_pith_number":"pith:75362SEH","schema_version":"1.0","canonical_sha256":"ff77ed488762291bcb8e967d40656b39d2de07dbb13ef4628335d078963a8f3f","source":{"kind":"arxiv","id":"2012.01536","version":3},"attestation_state":"computed","paper":{"title":"Improving KernelSHAP: Practical Shapley Value Estimation via Linear Regression","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Ian Covert, Su-In Lee","submitted_at":"2020-12-02T21:20:47Z","abstract_excerpt":"The Shapley value concept from cooperative game theory has become a popular technique for interpreting ML models, but efficiently estimating these values remains challenging, particularly in the model-agnostic setting. Here, we revisit the idea of estimating Shapley values via linear regression to understand and improve upon this approach. By analyzing the original KernelSHAP alongside a newly proposed unbiased version, we develop techniques to detect its convergence and calculate uncertainty estimates. We also find that the original version incurs a negligible increase in bias in exchange for"},"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":"2012.01536","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-12-02T21:20:47Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"fba34e6fd4860152862e91b236f899db3cd4582dafbad39993bdce74e264cab1","abstract_canon_sha256":"16a7659c93b3f2b60e686e8878b420768585cd3aaaf71778fe14bc6ded99d1cc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:34:20.973121Z","signature_b64":"ztV3XoxsaZWY4wamz0ybGG8M3Kk7EFNFJWgHBJILqaO5lO+LBA/plrL6k7g5fRx/WAgoVWYmyWUZGqZVk6kgBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ff77ed488762291bcb8e967d40656b39d2de07dbb13ef4628335d078963a8f3f","last_reissued_at":"2026-07-05T02:34:20.972670Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:34:20.972670Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving KernelSHAP: Practical Shapley Value Estimation via Linear Regression","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Ian Covert, Su-In Lee","submitted_at":"2020-12-02T21:20:47Z","abstract_excerpt":"The Shapley value concept from cooperative game theory has become a popular technique for interpreting ML models, but efficiently estimating these values remains challenging, particularly in the model-agnostic setting. Here, we revisit the idea of estimating Shapley values via linear regression to understand and improve upon this approach. By analyzing the original KernelSHAP alongside a newly proposed unbiased version, we develop techniques to detect its convergence and calculate uncertainty estimates. We also find that the original version incurs a negligible increase in bias in exchange for"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2012.01536","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/2012.01536/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":"2012.01536","created_at":"2026-07-05T02:34:20.972727+00:00"},{"alias_kind":"arxiv_version","alias_value":"2012.01536v3","created_at":"2026-07-05T02:34:20.972727+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2012.01536","created_at":"2026-07-05T02:34:20.972727+00:00"},{"alias_kind":"pith_short_12","alias_value":"75362SEHMIUR","created_at":"2026-07-05T02:34:20.972727+00:00"},{"alias_kind":"pith_short_16","alias_value":"75362SEHMIURXS4O","created_at":"2026-07-05T02:34:20.972727+00:00"},{"alias_kind":"pith_short_8","alias_value":"75362SEH","created_at":"2026-07-05T02:34:20.972727+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.05493","citing_title":"A renormalization-group inspired lattice-based framework for piecewise generalized linear models","ref_index":135,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/75362SEHMIURXS4OSZ6UAZLLHH","json":"https://pith.science/pith/75362SEHMIURXS4OSZ6UAZLLHH.json","graph_json":"https://pith.science/api/pith-number/75362SEHMIURXS4OSZ6UAZLLHH/graph.json","events_json":"https://pith.science/api/pith-number/75362SEHMIURXS4OSZ6UAZLLHH/events.json","paper":"https://pith.science/paper/75362SEH"},"agent_actions":{"view_html":"https://pith.science/pith/75362SEHMIURXS4OSZ6UAZLLHH","download_json":"https://pith.science/pith/75362SEHMIURXS4OSZ6UAZLLHH.json","view_paper":"https://pith.science/paper/75362SEH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2012.01536&json=true","fetch_graph":"https://pith.science/api/pith-number/75362SEHMIURXS4OSZ6UAZLLHH/graph.json","fetch_events":"https://pith.science/api/pith-number/75362SEHMIURXS4OSZ6UAZLLHH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/75362SEHMIURXS4OSZ6UAZLLHH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/75362SEHMIURXS4OSZ6UAZLLHH/action/storage_attestation","attest_author":"https://pith.science/pith/75362SEHMIURXS4OSZ6UAZLLHH/action/author_attestation","sign_citation":"https://pith.science/pith/75362SEHMIURXS4OSZ6UAZLLHH/action/citation_signature","submit_replication":"https://pith.science/pith/75362SEHMIURXS4OSZ6UAZLLHH/action/replication_record"}},"created_at":"2026-07-05T02:34:20.972727+00:00","updated_at":"2026-07-05T02:34:20.972727+00:00"}