{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2016:MY52NY7T4F4FEGK65SL6HDHLYR","short_pith_number":"pith:MY52NY7T","schema_version":"1.0","canonical_sha256":"663ba6e3f3e17852195eec97e38cebc4601ed7ae1b049fe0bb1d947dea12c1cf","source":{"kind":"arxiv","id":"1609.00150","version":3},"attestation_state":"computed","paper":{"title":"Reward Augmented Maximum Likelihood for Neural Structured Prediction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Dale Schuurmans, Mike Schuster, Mohammad Norouzi, Navdeep Jaitly, Samy Bengio, Yonghui Wu, Zhifeng Chen","submitted_at":"2016-09-01T09:00:19Z","abstract_excerpt":"A key problem in structured output prediction is direct optimization of the task reward function that matters for test evaluation. This paper presents a simple and computationally efficient approach to incorporate task reward into a maximum likelihood framework. By establishing a link between the log-likelihood and expected reward objectives, we show that an optimal regularized expected reward is achieved when the conditional distribution of the outputs given the inputs is proportional to their exponentiated scaled rewards. Accordingly, we present a framework to smooth the predictive probabili"},"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":"1609.00150","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-09-01T09:00:19Z","cross_cats_sorted":[],"title_canon_sha256":"9bc3df3ebea6fc593fd8ffbfb674186e90896c4d5dca6beb636393a3efee2cdb","abstract_canon_sha256":"17373e4f70f0abbaf8048bf5c7a5f403543dc20fe2e2f2bf8a020a60834080de"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:53:24.251417Z","signature_b64":"4gpoKa8/b/brmRciGpIsKAs2PQWKAq4a9ZAVQ4aev/1E58PXME2g5E8aeo9HU5RKpqht4kKwiVsvmwT8IdwfAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"663ba6e3f3e17852195eec97e38cebc4601ed7ae1b049fe0bb1d947dea12c1cf","last_reissued_at":"2026-05-18T00:53:24.250947Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:53:24.250947Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reward Augmented Maximum Likelihood for Neural Structured Prediction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Dale Schuurmans, Mike Schuster, Mohammad Norouzi, Navdeep Jaitly, Samy Bengio, Yonghui Wu, Zhifeng Chen","submitted_at":"2016-09-01T09:00:19Z","abstract_excerpt":"A key problem in structured output prediction is direct optimization of the task reward function that matters for test evaluation. This paper presents a simple and computationally efficient approach to incorporate task reward into a maximum likelihood framework. By establishing a link between the log-likelihood and expected reward objectives, we show that an optimal regularized expected reward is achieved when the conditional distribution of the outputs given the inputs is proportional to their exponentiated scaled rewards. Accordingly, we present a framework to smooth the predictive probabili"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1609.00150","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":""},"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":"1609.00150","created_at":"2026-05-18T00:53:24.251022+00:00"},{"alias_kind":"arxiv_version","alias_value":"1609.00150v3","created_at":"2026-05-18T00:53:24.251022+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1609.00150","created_at":"2026-05-18T00:53:24.251022+00:00"},{"alias_kind":"pith_short_12","alias_value":"MY52NY7T4F4F","created_at":"2026-05-18T12:30:32.724797+00:00"},{"alias_kind":"pith_short_16","alias_value":"MY52NY7T4F4FEGK6","created_at":"2026-05-18T12:30:32.724797+00:00"},{"alias_kind":"pith_short_8","alias_value":"MY52NY7T","created_at":"2026-05-18T12:30:32.724797+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MY52NY7T4F4FEGK65SL6HDHLYR","json":"https://pith.science/pith/MY52NY7T4F4FEGK65SL6HDHLYR.json","graph_json":"https://pith.science/api/pith-number/MY52NY7T4F4FEGK65SL6HDHLYR/graph.json","events_json":"https://pith.science/api/pith-number/MY52NY7T4F4FEGK65SL6HDHLYR/events.json","paper":"https://pith.science/paper/MY52NY7T"},"agent_actions":{"view_html":"https://pith.science/pith/MY52NY7T4F4FEGK65SL6HDHLYR","download_json":"https://pith.science/pith/MY52NY7T4F4FEGK65SL6HDHLYR.json","view_paper":"https://pith.science/paper/MY52NY7T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1609.00150&json=true","fetch_graph":"https://pith.science/api/pith-number/MY52NY7T4F4FEGK65SL6HDHLYR/graph.json","fetch_events":"https://pith.science/api/pith-number/MY52NY7T4F4FEGK65SL6HDHLYR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MY52NY7T4F4FEGK65SL6HDHLYR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MY52NY7T4F4FEGK65SL6HDHLYR/action/storage_attestation","attest_author":"https://pith.science/pith/MY52NY7T4F4FEGK65SL6HDHLYR/action/author_attestation","sign_citation":"https://pith.science/pith/MY52NY7T4F4FEGK65SL6HDHLYR/action/citation_signature","submit_replication":"https://pith.science/pith/MY52NY7T4F4FEGK65SL6HDHLYR/action/replication_record"}},"created_at":"2026-05-18T00:53:24.251022+00:00","updated_at":"2026-05-18T00:53:24.251022+00:00"}