{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:VX7MIYVHP2U7AFN34HIEAKPHSY","short_pith_number":"pith:VX7MIYVH","schema_version":"1.0","canonical_sha256":"adfec462a77ea9f015bbe1d04029e79623d1aa7dec7b4229ce7aa9239282ece4","source":{"kind":"arxiv","id":"2410.09133","version":2},"attestation_state":"computed","paper":{"title":"MVG-CRPS: A Robust Loss Function for Multivariate Probabilistic Forecasting","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Lijun Sun, Vincent Zhihao Zheng","submitted_at":"2024-10-11T15:10:20Z","abstract_excerpt":"Multivariate Gaussian (MVG) distributions are central to modeling correlated continuous variables in probabilistic forecasting. Neural forecasting models typically parameterize the mean vector and covariance matrix of the distribution using neural networks, optimizing with the log-score (negative log-likelihood) as the loss function. However, the sensitivity of the log-score to outliers can lead to significant errors in the presence of anomalies. Drawing on the continuous ranked probability score (CRPS) for univariate distributions, we propose MVG-CRPS, a strictly proper scoring rule for MVG d"},"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":"2410.09133","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2024-10-11T15:10:20Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"6d8635fe6c22ad64e223db3148de4a103f892f73ab291b5ce5765b2621a5570b","abstract_canon_sha256":"3dd12135ae5b93040c9ad44d664c2372d1496e28b1dd2da427e646fe6b9c14d5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:07:45.351983Z","signature_b64":"1KjbAbkdW0Oo3lMcfXxi2HDkb5f01sg53SejPO77LsLw2inXsVK7CMoX7WsvoNZkC3Rou9Vm9ufDewvqIVpuBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"adfec462a77ea9f015bbe1d04029e79623d1aa7dec7b4229ce7aa9239282ece4","last_reissued_at":"2026-07-05T10:07:45.351573Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:07:45.351573Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MVG-CRPS: A Robust Loss Function for Multivariate Probabilistic Forecasting","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Lijun Sun, Vincent Zhihao Zheng","submitted_at":"2024-10-11T15:10:20Z","abstract_excerpt":"Multivariate Gaussian (MVG) distributions are central to modeling correlated continuous variables in probabilistic forecasting. Neural forecasting models typically parameterize the mean vector and covariance matrix of the distribution using neural networks, optimizing with the log-score (negative log-likelihood) as the loss function. However, the sensitivity of the log-score to outliers can lead to significant errors in the presence of anomalies. Drawing on the continuous ranked probability score (CRPS) for univariate distributions, we propose MVG-CRPS, a strictly proper scoring rule for MVG d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.09133","kind":"arxiv","version":2},"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/2410.09133/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":"2410.09133","created_at":"2026-07-05T10:07:45.351626+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.09133v2","created_at":"2026-07-05T10:07:45.351626+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.09133","created_at":"2026-07-05T10:07:45.351626+00:00"},{"alias_kind":"pith_short_12","alias_value":"VX7MIYVHP2U7","created_at":"2026-07-05T10:07:45.351626+00:00"},{"alias_kind":"pith_short_16","alias_value":"VX7MIYVHP2U7AFN3","created_at":"2026-07-05T10:07:45.351626+00:00"},{"alias_kind":"pith_short_8","alias_value":"VX7MIYVH","created_at":"2026-07-05T10:07:45.351626+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.06473","citing_title":"MICA: Multivariate Infini Compressive Attention for Time Series Forecasting","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06473","citing_title":"MICA: Multivariate Infini Compressive Attention for Time Series Forecasting","ref_index":56,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VX7MIYVHP2U7AFN34HIEAKPHSY","json":"https://pith.science/pith/VX7MIYVHP2U7AFN34HIEAKPHSY.json","graph_json":"https://pith.science/api/pith-number/VX7MIYVHP2U7AFN34HIEAKPHSY/graph.json","events_json":"https://pith.science/api/pith-number/VX7MIYVHP2U7AFN34HIEAKPHSY/events.json","paper":"https://pith.science/paper/VX7MIYVH"},"agent_actions":{"view_html":"https://pith.science/pith/VX7MIYVHP2U7AFN34HIEAKPHSY","download_json":"https://pith.science/pith/VX7MIYVHP2U7AFN34HIEAKPHSY.json","view_paper":"https://pith.science/paper/VX7MIYVH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.09133&json=true","fetch_graph":"https://pith.science/api/pith-number/VX7MIYVHP2U7AFN34HIEAKPHSY/graph.json","fetch_events":"https://pith.science/api/pith-number/VX7MIYVHP2U7AFN34HIEAKPHSY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VX7MIYVHP2U7AFN34HIEAKPHSY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VX7MIYVHP2U7AFN34HIEAKPHSY/action/storage_attestation","attest_author":"https://pith.science/pith/VX7MIYVHP2U7AFN34HIEAKPHSY/action/author_attestation","sign_citation":"https://pith.science/pith/VX7MIYVHP2U7AFN34HIEAKPHSY/action/citation_signature","submit_replication":"https://pith.science/pith/VX7MIYVHP2U7AFN34HIEAKPHSY/action/replication_record"}},"created_at":"2026-07-05T10:07:45.351626+00:00","updated_at":"2026-07-05T10:07:45.351626+00:00"}