{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:7CQ5WH7JVQFF7GXSM3NFI4CPVZ","short_pith_number":"pith:7CQ5WH7J","schema_version":"1.0","canonical_sha256":"f8a1db1fe9ac0a5f9af266da54704fae45592bd682f292ba11c49b978e662dd6","source":{"kind":"arxiv","id":"2509.08369","version":1},"attestation_state":"computed","paper":{"title":"Stop using root-mean-square error as a precipitation target!","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"physics.ao-ph","authors_text":"Kieran M. R. Hunt","submitted_at":"2025-09-10T08:06:24Z","abstract_excerpt":"Root-mean-square error (RMSE) remains the default training loss for data-driven precipitation models, despite precipitation being semi-continuous, zero-inflated, strictly non-negative, and heavy-tailed. This Gaussian-implied objective misspecifies the data-generating process because it tolerates negative predictions, underpenalises rare heavy events, and ignores the mass at zero. We propose replacing RMSE with the Tweedie deviance, a likelihood-based and differentiable loss from the exponential--dispersion family with variance function $V(\\mu)=\\mu^p$. For $1<p<2$ it yields a compound Poisson--"},"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":"2509.08369","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"physics.ao-ph","submitted_at":"2025-09-10T08:06:24Z","cross_cats_sorted":[],"title_canon_sha256":"42a7ca7bb69dd2af02c900aab322133cabb175ff12c52f1faac2cec98bff926a","abstract_canon_sha256":"950e47d31ebfd10bbd1a63aa5ae1981b6276e06f404cdc7eec3add79bf5f523f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:08:28.767517Z","signature_b64":"FHxMzpwMeDzkJVAFh496JI/zVdSQx3qmpMGhx7ZTc/zuqipwEZOE/W16Er8O2v+wQbXCqS7+aGjf9e9W0luPDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f8a1db1fe9ac0a5f9af266da54704fae45592bd682f292ba11c49b978e662dd6","last_reissued_at":"2026-07-05T12:08:28.766985Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:08:28.766985Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Stop using root-mean-square error as a precipitation target!","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"physics.ao-ph","authors_text":"Kieran M. R. Hunt","submitted_at":"2025-09-10T08:06:24Z","abstract_excerpt":"Root-mean-square error (RMSE) remains the default training loss for data-driven precipitation models, despite precipitation being semi-continuous, zero-inflated, strictly non-negative, and heavy-tailed. This Gaussian-implied objective misspecifies the data-generating process because it tolerates negative predictions, underpenalises rare heavy events, and ignores the mass at zero. We propose replacing RMSE with the Tweedie deviance, a likelihood-based and differentiable loss from the exponential--dispersion family with variance function $V(\\mu)=\\mu^p$. For $1<p<2$ it yields a compound Poisson--"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.08369","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/2509.08369/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":"2509.08369","created_at":"2026-07-05T12:08:28.767053+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.08369v1","created_at":"2026-07-05T12:08:28.767053+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.08369","created_at":"2026-07-05T12:08:28.767053+00:00"},{"alias_kind":"pith_short_12","alias_value":"7CQ5WH7JVQFF","created_at":"2026-07-05T12:08:28.767053+00:00"},{"alias_kind":"pith_short_16","alias_value":"7CQ5WH7JVQFF7GXS","created_at":"2026-07-05T12:08:28.767053+00:00"},{"alias_kind":"pith_short_8","alias_value":"7CQ5WH7J","created_at":"2026-07-05T12:08:28.767053+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.05854","citing_title":"AirQualityBench: A Realistic Evaluation Benchmark for Global Air Quality Forecasting","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7CQ5WH7JVQFF7GXSM3NFI4CPVZ","json":"https://pith.science/pith/7CQ5WH7JVQFF7GXSM3NFI4CPVZ.json","graph_json":"https://pith.science/api/pith-number/7CQ5WH7JVQFF7GXSM3NFI4CPVZ/graph.json","events_json":"https://pith.science/api/pith-number/7CQ5WH7JVQFF7GXSM3NFI4CPVZ/events.json","paper":"https://pith.science/paper/7CQ5WH7J"},"agent_actions":{"view_html":"https://pith.science/pith/7CQ5WH7JVQFF7GXSM3NFI4CPVZ","download_json":"https://pith.science/pith/7CQ5WH7JVQFF7GXSM3NFI4CPVZ.json","view_paper":"https://pith.science/paper/7CQ5WH7J","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.08369&json=true","fetch_graph":"https://pith.science/api/pith-number/7CQ5WH7JVQFF7GXSM3NFI4CPVZ/graph.json","fetch_events":"https://pith.science/api/pith-number/7CQ5WH7JVQFF7GXSM3NFI4CPVZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7CQ5WH7JVQFF7GXSM3NFI4CPVZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7CQ5WH7JVQFF7GXSM3NFI4CPVZ/action/storage_attestation","attest_author":"https://pith.science/pith/7CQ5WH7JVQFF7GXSM3NFI4CPVZ/action/author_attestation","sign_citation":"https://pith.science/pith/7CQ5WH7JVQFF7GXSM3NFI4CPVZ/action/citation_signature","submit_replication":"https://pith.science/pith/7CQ5WH7JVQFF7GXSM3NFI4CPVZ/action/replication_record"}},"created_at":"2026-07-05T12:08:28.767053+00:00","updated_at":"2026-07-05T12:08:28.767053+00:00"}