{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:YSFC2RUEAESIBIOQTTPH3LRJMO","short_pith_number":"pith:YSFC2RUE","canonical_record":{"source":{"id":"2401.14483","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-01-25T19:36:11Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"7a58b0044b42a04a3ff8f57895e2d13b30a76fa0148205089b8238b2ec644027","abstract_canon_sha256":"f1422896a73d4b28606b2792ce3bcbad79abc38fdf72474b8d9f13a26054ef58"},"schema_version":"1.0"},"canonical_sha256":"c48a2d4684012480a1d09cde7dae2963b1f80cc7e26ce7784148f7195c1702ba","source":{"kind":"arxiv","id":"2401.14483","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.14483","created_at":"2026-06-24T01:14:55Z"},{"alias_kind":"arxiv_version","alias_value":"2401.14483v4","created_at":"2026-06-24T01:14:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.14483","created_at":"2026-06-24T01:14:55Z"},{"alias_kind":"pith_short_12","alias_value":"YSFC2RUEAESI","created_at":"2026-06-24T01:14:55Z"},{"alias_kind":"pith_short_16","alias_value":"YSFC2RUEAESIBIOQ","created_at":"2026-06-24T01:14:55Z"},{"alias_kind":"pith_short_8","alias_value":"YSFC2RUE","created_at":"2026-06-24T01:14:55Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:YSFC2RUEAESIBIOQTTPH3LRJMO","target":"record","payload":{"canonical_record":{"source":{"id":"2401.14483","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-01-25T19:36:11Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"7a58b0044b42a04a3ff8f57895e2d13b30a76fa0148205089b8238b2ec644027","abstract_canon_sha256":"f1422896a73d4b28606b2792ce3bcbad79abc38fdf72474b8d9f13a26054ef58"},"schema_version":"1.0"},"canonical_sha256":"c48a2d4684012480a1d09cde7dae2963b1f80cc7e26ce7784148f7195c1702ba","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-24T01:14:55.197916Z","signature_b64":"QhhX45jzAjICL+WCXjNr+7Yqu6NJGgPbKWvc+bLxot/S69t2BIJHFIcNqIJX7K6SsfMClfJjExnOrxpl5nCNBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c48a2d4684012480a1d09cde7dae2963b1f80cc7e26ce7784148f7195c1702ba","last_reissued_at":"2026-06-24T01:14:55.197423Z","signature_status":"signed_v1","first_computed_at":"2026-06-24T01:14:55.197423Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2401.14483","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-24T01:14:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZQDirEJW5VBAT9ddISze82kEKBe3b01DE0+NT0iGGoIOixAH1U6KeVJjDYb+MZR26CXnRpaAEiAMkcy4kDHODQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-27T07:44:28.410876Z"},"content_sha256":"5e8278464e13bf4dd4e9456b7e6cec75600d80f79f9b60911153574121787eef","schema_version":"1.0","event_id":"sha256:5e8278464e13bf4dd4e9456b7e6cec75600d80f79f9b60911153574121787eef"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:YSFC2RUEAESIBIOQTTPH3LRJMO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Evaluation Metrics as Averaged Outcomes of Fair Gambles","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Rabanus Derr, Robert C. Williamson","submitted_at":"2024-01-25T19:36:11Z","abstract_excerpt":"In the current practices of machine learning, the evaluation of forecasts has become a cornerstone of scientific progress. A multitude of evaluation metrics have been suggested and used to qualify \"good\" forecasts. What do those metrics share? How are they related? In this work, we use a protocol borrowed from game-theoretic probability to show that a large part of evaluation metrics can be viewed as averaged outcomes of fair gambles. Intuitively, a fair gambler is one which a forecaster would expect to fail. Hence, the gambler's ability to gain disproves the quality of the forecast. Standard "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.14483","kind":"arxiv","version":4},"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/2401.14483/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-24T01:14:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YaFeLFSeVk8TsFwHrG6E5g3Nrh15rqCatKEfuL7wOPKHx3/6R3pFXOLzNKb1eL0PDnJZBKV/TOaF6S2p0feEAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-27T07:44:28.411242Z"},"content_sha256":"3297e0823bfd7b2b364333453e68f424fb6637aeacb241f71b02ecbdebba79c7","schema_version":"1.0","event_id":"sha256:3297e0823bfd7b2b364333453e68f424fb6637aeacb241f71b02ecbdebba79c7"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YSFC2RUEAESIBIOQTTPH3LRJMO/bundle.json","state_url":"https://pith.science/pith/YSFC2RUEAESIBIOQTTPH3LRJMO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YSFC2RUEAESIBIOQTTPH3LRJMO/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-27T07:44:28Z","links":{"resolver":"https://pith.science/pith/YSFC2RUEAESIBIOQTTPH3LRJMO","bundle":"https://pith.science/pith/YSFC2RUEAESIBIOQTTPH3LRJMO/bundle.json","state":"https://pith.science/pith/YSFC2RUEAESIBIOQTTPH3LRJMO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YSFC2RUEAESIBIOQTTPH3LRJMO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:YSFC2RUEAESIBIOQTTPH3LRJMO","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"f1422896a73d4b28606b2792ce3bcbad79abc38fdf72474b8d9f13a26054ef58","cross_cats_sorted":["stat.ML"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-01-25T19:36:11Z","title_canon_sha256":"7a58b0044b42a04a3ff8f57895e2d13b30a76fa0148205089b8238b2ec644027"},"schema_version":"1.0","source":{"id":"2401.14483","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.14483","created_at":"2026-06-24T01:14:55Z"},{"alias_kind":"arxiv_version","alias_value":"2401.14483v4","created_at":"2026-06-24T01:14:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.14483","created_at":"2026-06-24T01:14:55Z"},{"alias_kind":"pith_short_12","alias_value":"YSFC2RUEAESI","created_at":"2026-06-24T01:14:55Z"},{"alias_kind":"pith_short_16","alias_value":"YSFC2RUEAESIBIOQ","created_at":"2026-06-24T01:14:55Z"},{"alias_kind":"pith_short_8","alias_value":"YSFC2RUE","created_at":"2026-06-24T01:14:55Z"}],"graph_snapshots":[{"event_id":"sha256:3297e0823bfd7b2b364333453e68f424fb6637aeacb241f71b02ecbdebba79c7","target":"graph","created_at":"2026-06-24T01:14:55Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2401.14483/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In the current practices of machine learning, the evaluation of forecasts has become a cornerstone of scientific progress. A multitude of evaluation metrics have been suggested and used to qualify \"good\" forecasts. What do those metrics share? How are they related? In this work, we use a protocol borrowed from game-theoretic probability to show that a large part of evaluation metrics can be viewed as averaged outcomes of fair gambles. Intuitively, a fair gambler is one which a forecaster would expect to fail. Hence, the gambler's ability to gain disproves the quality of the forecast. Standard ","authors_text":"Rabanus Derr, Robert C. Williamson","cross_cats":["stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-01-25T19:36:11Z","title":"Evaluation Metrics as Averaged Outcomes of Fair Gambles"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.14483","kind":"arxiv","version":4},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:5e8278464e13bf4dd4e9456b7e6cec75600d80f79f9b60911153574121787eef","target":"record","created_at":"2026-06-24T01:14:55Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"f1422896a73d4b28606b2792ce3bcbad79abc38fdf72474b8d9f13a26054ef58","cross_cats_sorted":["stat.ML"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-01-25T19:36:11Z","title_canon_sha256":"7a58b0044b42a04a3ff8f57895e2d13b30a76fa0148205089b8238b2ec644027"},"schema_version":"1.0","source":{"id":"2401.14483","kind":"arxiv","version":4}},"canonical_sha256":"c48a2d4684012480a1d09cde7dae2963b1f80cc7e26ce7784148f7195c1702ba","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c48a2d4684012480a1d09cde7dae2963b1f80cc7e26ce7784148f7195c1702ba","first_computed_at":"2026-06-24T01:14:55.197423Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-24T01:14:55.197423Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"QhhX45jzAjICL+WCXjNr+7Yqu6NJGgPbKWvc+bLxot/S69t2BIJHFIcNqIJX7K6SsfMClfJjExnOrxpl5nCNBw==","signature_status":"signed_v1","signed_at":"2026-06-24T01:14:55.197916Z","signed_message":"canonical_sha256_bytes"},"source_id":"2401.14483","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5e8278464e13bf4dd4e9456b7e6cec75600d80f79f9b60911153574121787eef","sha256:3297e0823bfd7b2b364333453e68f424fb6637aeacb241f71b02ecbdebba79c7"],"state_sha256":"204acc1f0e5d43c7fb3792b0a540c609e4291bac0f4a2d520bbb63c5e3a03397"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OP2L/sUyzQsEE8oDaG8cO1gb0n7/siLe0MpYDvyZQeMBfHDKh7Nwn3BRc1Y+DIHpv8cOpZi1U0MtGdLvIgpRDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-27T07:44:28.414208Z","bundle_sha256":"a2f456ceb1dd0948729a1f70ed30577580dedb5164dc621e4ee34e4eca932257"}}