{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:T6UWNQTS4PCASEAHGIHWKNLAIJ","short_pith_number":"pith:T6UWNQTS","schema_version":"1.0","canonical_sha256":"9fa966c272e3c4091007320f65356042465f6c7a533ee8bcd7ef8b515704c923","source":{"kind":"arxiv","id":"2112.07535","version":2},"attestation_state":"computed","paper":{"title":"Scientific Discovery and the Cost of Measurement -- Balancing Information and Cost in Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Andriy Drozdyuk, Colin Bellinger, Isaac Tamblyn, Mark Crowley","submitted_at":"2021-12-14T16:48:57Z","abstract_excerpt":"The use of reinforcement learning (RL) in scientific applications, such as materials design and automated chemistry, is increasing. A major challenge, however, lies in fact that measuring the state of the system is often costly and time consuming in scientific applications, whereas policy learning with RL requires a measurement after each time step. In this work, we make the measurement costs explicit in the form of a costed reward and propose a framework that enables off-the-shelf deep RL algorithms to learn a policy for both selecting actions and determining whether or not to measure the cur"},"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":"2112.07535","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-12-14T16:48:57Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a7e944a2b500a30bb9356f2daff688ec6289eef021f691ad2eab9490c99503e9","abstract_canon_sha256":"b21a87e1908fa1cf8adb2238d92e4502b4ed301b32a6486b244db7f200a7fcf1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:12:17.338869Z","signature_b64":"LiC+FCTT7i2IGFZf+owCkGV9WwA9abb1iZCJ50xuN1yAGESQJoW8EK2tTSpPw/4ft/wlu959vT4rvZQSSSAdCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9fa966c272e3c4091007320f65356042465f6c7a533ee8bcd7ef8b515704c923","last_reissued_at":"2026-07-05T04:12:17.338390Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:12:17.338390Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scientific Discovery and the Cost of Measurement -- Balancing Information and Cost in Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Andriy Drozdyuk, Colin Bellinger, Isaac Tamblyn, Mark Crowley","submitted_at":"2021-12-14T16:48:57Z","abstract_excerpt":"The use of reinforcement learning (RL) in scientific applications, such as materials design and automated chemistry, is increasing. A major challenge, however, lies in fact that measuring the state of the system is often costly and time consuming in scientific applications, whereas policy learning with RL requires a measurement after each time step. In this work, we make the measurement costs explicit in the form of a costed reward and propose a framework that enables off-the-shelf deep RL algorithms to learn a policy for both selecting actions and determining whether or not to measure the cur"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.07535","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/2112.07535/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":"2112.07535","created_at":"2026-07-05T04:12:17.338443+00:00"},{"alias_kind":"arxiv_version","alias_value":"2112.07535v2","created_at":"2026-07-05T04:12:17.338443+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.07535","created_at":"2026-07-05T04:12:17.338443+00:00"},{"alias_kind":"pith_short_12","alias_value":"T6UWNQTS4PCA","created_at":"2026-07-05T04:12:17.338443+00:00"},{"alias_kind":"pith_short_16","alias_value":"T6UWNQTS4PCASEAH","created_at":"2026-07-05T04:12:17.338443+00:00"},{"alias_kind":"pith_short_8","alias_value":"T6UWNQTS","created_at":"2026-07-05T04:12:17.338443+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.12823","citing_title":"To Measure or Not: A Cost-Sensitive, Selective Measuring Environment for Agricultural Management Decisions with Reinforcement Learning","ref_index":4,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/T6UWNQTS4PCASEAHGIHWKNLAIJ","json":"https://pith.science/pith/T6UWNQTS4PCASEAHGIHWKNLAIJ.json","graph_json":"https://pith.science/api/pith-number/T6UWNQTS4PCASEAHGIHWKNLAIJ/graph.json","events_json":"https://pith.science/api/pith-number/T6UWNQTS4PCASEAHGIHWKNLAIJ/events.json","paper":"https://pith.science/paper/T6UWNQTS"},"agent_actions":{"view_html":"https://pith.science/pith/T6UWNQTS4PCASEAHGIHWKNLAIJ","download_json":"https://pith.science/pith/T6UWNQTS4PCASEAHGIHWKNLAIJ.json","view_paper":"https://pith.science/paper/T6UWNQTS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2112.07535&json=true","fetch_graph":"https://pith.science/api/pith-number/T6UWNQTS4PCASEAHGIHWKNLAIJ/graph.json","fetch_events":"https://pith.science/api/pith-number/T6UWNQTS4PCASEAHGIHWKNLAIJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/T6UWNQTS4PCASEAHGIHWKNLAIJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/T6UWNQTS4PCASEAHGIHWKNLAIJ/action/storage_attestation","attest_author":"https://pith.science/pith/T6UWNQTS4PCASEAHGIHWKNLAIJ/action/author_attestation","sign_citation":"https://pith.science/pith/T6UWNQTS4PCASEAHGIHWKNLAIJ/action/citation_signature","submit_replication":"https://pith.science/pith/T6UWNQTS4PCASEAHGIHWKNLAIJ/action/replication_record"}},"created_at":"2026-07-05T04:12:17.338443+00:00","updated_at":"2026-07-05T04:12:17.338443+00:00"}