{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:RV5ZVD22IJXAAP3464CJ7VEFC6","short_pith_number":"pith:RV5ZVD22","schema_version":"1.0","canonical_sha256":"8d7b9a8f5a426e003f7cf7049fd48517a0f06b2340b363b0a066123bb5e175dd","source":{"kind":"arxiv","id":"2012.15511","version":2},"attestation_state":"computed","paper":{"title":"Towards Understanding Asynchronous Advantage Actor-critic: Convergence and Linear Speedup","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC"],"primary_cat":"cs.LG","authors_text":"Han Shen, Kaiqing Zhang, Mingyi Hong, Tianyi Chen","submitted_at":"2020-12-31T09:07:09Z","abstract_excerpt":"Asynchronous and parallel implementation of standard reinforcement learning (RL) algorithms is a key enabler of the tremendous success of modern RL. Among many asynchronous RL algorithms, arguably the most popular and effective one is the asynchronous advantage actor-critic (A3C) algorithm. Although A3C is becoming the workhorse of RL, its theoretical properties are still not well-understood, including its non-asymptotic analysis and the performance gain of parallelism (a.k.a. linear speedup). This paper revisits the A3C algorithm and establishes its non-asymptotic convergence guarantees. Unde"},"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":"2012.15511","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-12-31T09:07:09Z","cross_cats_sorted":["math.OC"],"title_canon_sha256":"d5260aa4a899ca6da0f1aa72f6c60f61efd69e4e44fd6986fffc37d979a819ef","abstract_canon_sha256":"e302d15b0658cee29dd7511ec2a3cd4ba84068b908bdc6c2a499900744199137"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:36:18.767307Z","signature_b64":"vKsesLQliDNTheg8daP5QX92y8eBIeUKIovoJICmBl399HbmclNnpZgMkoCAcakE8jeFHgiKCy7Js4rL3vPvDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8d7b9a8f5a426e003f7cf7049fd48517a0f06b2340b363b0a066123bb5e175dd","last_reissued_at":"2026-07-05T06:36:18.766851Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:36:18.766851Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Understanding Asynchronous Advantage Actor-critic: Convergence and Linear Speedup","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC"],"primary_cat":"cs.LG","authors_text":"Han Shen, Kaiqing Zhang, Mingyi Hong, Tianyi Chen","submitted_at":"2020-12-31T09:07:09Z","abstract_excerpt":"Asynchronous and parallel implementation of standard reinforcement learning (RL) algorithms is a key enabler of the tremendous success of modern RL. Among many asynchronous RL algorithms, arguably the most popular and effective one is the asynchronous advantage actor-critic (A3C) algorithm. Although A3C is becoming the workhorse of RL, its theoretical properties are still not well-understood, including its non-asymptotic analysis and the performance gain of parallelism (a.k.a. linear speedup). This paper revisits the A3C algorithm and establishes its non-asymptotic convergence guarantees. Unde"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2012.15511","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/2012.15511/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":"2012.15511","created_at":"2026-07-05T06:36:18.766910+00:00"},{"alias_kind":"arxiv_version","alias_value":"2012.15511v2","created_at":"2026-07-05T06:36:18.766910+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2012.15511","created_at":"2026-07-05T06:36:18.766910+00:00"},{"alias_kind":"pith_short_12","alias_value":"RV5ZVD22IJXA","created_at":"2026-07-05T06:36:18.766910+00:00"},{"alias_kind":"pith_short_16","alias_value":"RV5ZVD22IJXAAP34","created_at":"2026-07-05T06:36:18.766910+00:00"},{"alias_kind":"pith_short_8","alias_value":"RV5ZVD22","created_at":"2026-07-05T06:36:18.766910+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.18299","citing_title":"Model-Free RL Agents Demonstrate System 1-Like Intentionality","ref_index":49,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RV5ZVD22IJXAAP3464CJ7VEFC6","json":"https://pith.science/pith/RV5ZVD22IJXAAP3464CJ7VEFC6.json","graph_json":"https://pith.science/api/pith-number/RV5ZVD22IJXAAP3464CJ7VEFC6/graph.json","events_json":"https://pith.science/api/pith-number/RV5ZVD22IJXAAP3464CJ7VEFC6/events.json","paper":"https://pith.science/paper/RV5ZVD22"},"agent_actions":{"view_html":"https://pith.science/pith/RV5ZVD22IJXAAP3464CJ7VEFC6","download_json":"https://pith.science/pith/RV5ZVD22IJXAAP3464CJ7VEFC6.json","view_paper":"https://pith.science/paper/RV5ZVD22","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2012.15511&json=true","fetch_graph":"https://pith.science/api/pith-number/RV5ZVD22IJXAAP3464CJ7VEFC6/graph.json","fetch_events":"https://pith.science/api/pith-number/RV5ZVD22IJXAAP3464CJ7VEFC6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RV5ZVD22IJXAAP3464CJ7VEFC6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RV5ZVD22IJXAAP3464CJ7VEFC6/action/storage_attestation","attest_author":"https://pith.science/pith/RV5ZVD22IJXAAP3464CJ7VEFC6/action/author_attestation","sign_citation":"https://pith.science/pith/RV5ZVD22IJXAAP3464CJ7VEFC6/action/citation_signature","submit_replication":"https://pith.science/pith/RV5ZVD22IJXAAP3464CJ7VEFC6/action/replication_record"}},"created_at":"2026-07-05T06:36:18.766910+00:00","updated_at":"2026-07-05T06:36:18.766910+00:00"}