{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6THIRS4Q37EIGOF45EXIY35MUX","short_pith_number":"pith:6THIRS4Q","schema_version":"1.0","canonical_sha256":"f4ce88cb90dfc88338bce92e8c6faca5f6407f48c2bc4e89de5e91eeb88b1388","source":{"kind":"arxiv","id":"2605.26418","version":1},"attestation_state":"computed","paper":{"title":"When Does Deep RL Beat Calibrated Baselines? A Benchmark Study on Adaptive Resource Control","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.DC"],"primary_cat":"cs.LG","authors_text":"Chuanyi Sun, Guilin Zhang, John Fossaceca, Kai Zhao, Shahryar Sarkani","submitted_at":"2026-05-26T01:07:42Z","abstract_excerpt":"A properly calibrated rule-based autoscaler can beat every one of six mainstream deep reinforcement learning (DRL) algorithms on cost across every workload we test - so when, if ever, does DRL actually help? We study this in RLScale-Bench, a reproducible benchmark and evaluation protocol for DRL on adaptive resource control, where an agent allocates compute to a dynamic workload under cost and service-level constraints. We evaluate PPO, DQN, A2C, SAC, TD3, and DDPG under matched architectures, training budgets, and reward functions against a calibrated rule-based baseline across six workload p"},"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":"2605.26418","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-26T01:07:42Z","cross_cats_sorted":["cs.AI","cs.DC"],"title_canon_sha256":"7300f6bbbad80719e89e30fac9bb0469dc9bbbbe72e883a017d70fe7f179620d","abstract_canon_sha256":"da645aa94f765053e87e9ac10d1e59010d6ec9e412988f61dafe435eeb7d9a48"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-27T01:05:16.859210Z","signature_b64":"N2/TuEiTFxGXKxpXBjxPP0gmVN/g5sjOHQ+N/69eT0pSW+XK3DafSfaI/HhGEhlNYZyE9ZinoO7BfPFJgvpUAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f4ce88cb90dfc88338bce92e8c6faca5f6407f48c2bc4e89de5e91eeb88b1388","last_reissued_at":"2026-05-27T01:05:16.858522Z","signature_status":"signed_v1","first_computed_at":"2026-05-27T01:05:16.858522Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"When Does Deep RL Beat Calibrated Baselines? A Benchmark Study on Adaptive Resource Control","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.DC"],"primary_cat":"cs.LG","authors_text":"Chuanyi Sun, Guilin Zhang, John Fossaceca, Kai Zhao, Shahryar Sarkani","submitted_at":"2026-05-26T01:07:42Z","abstract_excerpt":"A properly calibrated rule-based autoscaler can beat every one of six mainstream deep reinforcement learning (DRL) algorithms on cost across every workload we test - so when, if ever, does DRL actually help? We study this in RLScale-Bench, a reproducible benchmark and evaluation protocol for DRL on adaptive resource control, where an agent allocates compute to a dynamic workload under cost and service-level constraints. We evaluate PPO, DQN, A2C, SAC, TD3, and DDPG under matched architectures, training budgets, and reward functions against a calibrated rule-based baseline across six workload p"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.26418","kind":"arxiv","version":1},"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/2605.26418/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":"2605.26418","created_at":"2026-05-27T01:05:16.858645+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.26418v1","created_at":"2026-05-27T01:05:16.858645+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.26418","created_at":"2026-05-27T01:05:16.858645+00:00"},{"alias_kind":"pith_short_12","alias_value":"6THIRS4Q37EI","created_at":"2026-05-27T01:05:16.858645+00:00"},{"alias_kind":"pith_short_16","alias_value":"6THIRS4Q37EIGOF4","created_at":"2026-05-27T01:05:16.858645+00:00"},{"alias_kind":"pith_short_8","alias_value":"6THIRS4Q","created_at":"2026-05-27T01:05:16.858645+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/6THIRS4Q37EIGOF45EXIY35MUX","json":"https://pith.science/pith/6THIRS4Q37EIGOF45EXIY35MUX.json","graph_json":"https://pith.science/api/pith-number/6THIRS4Q37EIGOF45EXIY35MUX/graph.json","events_json":"https://pith.science/api/pith-number/6THIRS4Q37EIGOF45EXIY35MUX/events.json","paper":"https://pith.science/paper/6THIRS4Q"},"agent_actions":{"view_html":"https://pith.science/pith/6THIRS4Q37EIGOF45EXIY35MUX","download_json":"https://pith.science/pith/6THIRS4Q37EIGOF45EXIY35MUX.json","view_paper":"https://pith.science/paper/6THIRS4Q","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.26418&json=true","fetch_graph":"https://pith.science/api/pith-number/6THIRS4Q37EIGOF45EXIY35MUX/graph.json","fetch_events":"https://pith.science/api/pith-number/6THIRS4Q37EIGOF45EXIY35MUX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6THIRS4Q37EIGOF45EXIY35MUX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6THIRS4Q37EIGOF45EXIY35MUX/action/storage_attestation","attest_author":"https://pith.science/pith/6THIRS4Q37EIGOF45EXIY35MUX/action/author_attestation","sign_citation":"https://pith.science/pith/6THIRS4Q37EIGOF45EXIY35MUX/action/citation_signature","submit_replication":"https://pith.science/pith/6THIRS4Q37EIGOF45EXIY35MUX/action/replication_record"}},"created_at":"2026-05-27T01:05:16.858645+00:00","updated_at":"2026-05-27T01:05:16.858645+00:00"}