{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:P4OWRCKXSZ7GSJ3A7LQLJ7VHDG","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":"055f29d3673275baae2d2ff13cae2df3778113e4b0ba33cd6008282f40fcec74","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-22T06:43:51Z","title_canon_sha256":"e10f0fc8494fb65599e5fedfe2c211ac9e1ead1daaeefaa45791942b579952b5"},"schema_version":"1.0","source":{"id":"2406.15753","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.15753","created_at":"2026-07-05T11:33:15Z"},{"alias_kind":"arxiv_version","alias_value":"2406.15753v3","created_at":"2026-07-05T11:33:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.15753","created_at":"2026-07-05T11:33:15Z"},{"alias_kind":"pith_short_12","alias_value":"P4OWRCKXSZ7G","created_at":"2026-07-05T11:33:15Z"},{"alias_kind":"pith_short_16","alias_value":"P4OWRCKXSZ7GSJ3A","created_at":"2026-07-05T11:33:15Z"},{"alias_kind":"pith_short_8","alias_value":"P4OWRCKX","created_at":"2026-07-05T11:33:15Z"}],"graph_snapshots":[{"event_id":"sha256:f613de6b8e716b6299d611abd6efece06457b2a7fcea39370150942eea3f2241","target":"graph","created_at":"2026-07-05T11:33:15Z","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/2406.15753/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In reinforcement learning, specifying reward functions that capture the intended task can be very challenging. Reward learning aims to address this issue by learning the reward function. However, a learned reward model may have a low error on the data distribution, and yet subsequently produce a policy with large regret. We say that such a reward model has an error-regret mismatch. The main source of an error-regret mismatch is the distributional shift that commonly occurs during policy optimization. In this paper, we mathematically show that a sufficiently low expected test error of the rewar","authors_text":"Alessandro Abate, David Krueger, Joar Skalse, Leon Lang, Lukas Fluri, Patrick Forr\\'e","cross_cats":["cs.AI","stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-22T06:43:51Z","title":"The Perils of Optimizing Learned Reward Functions: Low Training Error Does Not Guarantee Low Regret"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.15753","kind":"arxiv","version":3},"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":"sha256:0743854160798d92096a51f13f20553723f4fefd12bf768111a9c717c5891580","target":"record","created_at":"2026-07-05T11:33:15Z","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":"055f29d3673275baae2d2ff13cae2df3778113e4b0ba33cd6008282f40fcec74","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-22T06:43:51Z","title_canon_sha256":"e10f0fc8494fb65599e5fedfe2c211ac9e1ead1daaeefaa45791942b579952b5"},"schema_version":"1.0","source":{"id":"2406.15753","kind":"arxiv","version":3}},"canonical_sha256":"7f1d688957967e692760fae0b4fea7199789a3c7a7c19bc798008b3de5230be7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7f1d688957967e692760fae0b4fea7199789a3c7a7c19bc798008b3de5230be7","first_computed_at":"2026-07-05T11:33:15.052286Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:33:15.052286Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"63rb7R+yImUkH/D6TMpihcIN6GmIsgmlag9cMw1c+Rx+6WZrMiWz43h+yot5aaJmmeXKoNcCck+Fk1jtxuifDg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:33:15.052788Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.15753","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0743854160798d92096a51f13f20553723f4fefd12bf768111a9c717c5891580","sha256:f613de6b8e716b6299d611abd6efece06457b2a7fcea39370150942eea3f2241"],"state_sha256":"ad14da8802dec3ae5d0b5ae64d3ce4a855966f36612b6df840a7a5cfd50d516c"}