{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:NK3FHRXT6CKKJ7LWV3BM4QHEQ2","short_pith_number":"pith:NK3FHRXT","schema_version":"1.0","canonical_sha256":"6ab653c6f3f094a4fd76aec2ce40e486a6512b849abab0f40dc2f9f3367b267f","source":{"kind":"arxiv","id":"1803.02348","version":3},"attestation_state":"computed","paper":{"title":"Smoothed Action Value Functions for Learning Gaussian Policies","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Dale Schuurmans, George Tucker, Mohammad Norouzi, Ofir Nachum","submitted_at":"2018-03-06T04:58:20Z","abstract_excerpt":"State-action value functions (i.e., Q-values) are ubiquitous in reinforcement learning (RL), giving rise to popular algorithms such as SARSA and Q-learning. We propose a new notion of action value defined by a Gaussian smoothed version of the expected Q-value. We show that such smoothed Q-values still satisfy a Bellman equation, making them learnable from experience sampled from an environment. Moreover, the gradients of expected reward with respect to the mean and covariance of a parameterized Gaussian policy can be recovered from the gradient and Hessian of the smoothed Q-value function. Bas"},"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":"1803.02348","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-03-06T04:58:20Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"0029b13a3dbaea203082c99e58ff30dee472d201f61d2e92085f8302f5ac2d9a","abstract_canon_sha256":"983cef737b491c9d231a7309d438f07b99073ea6eb404b98e4c6343c35db405c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:09:52.558370Z","signature_b64":"x00fPR4NnOLH5C1HkR9CeCI2j0l25y4C6FocEPnzS5nnRYZy0QUaHyX05W8RPS4G1wuWHWL98SkMd/bGb69iAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6ab653c6f3f094a4fd76aec2ce40e486a6512b849abab0f40dc2f9f3367b267f","last_reissued_at":"2026-05-18T00:09:52.557688Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:09:52.557688Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Smoothed Action Value Functions for Learning Gaussian Policies","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Dale Schuurmans, George Tucker, Mohammad Norouzi, Ofir Nachum","submitted_at":"2018-03-06T04:58:20Z","abstract_excerpt":"State-action value functions (i.e., Q-values) are ubiquitous in reinforcement learning (RL), giving rise to popular algorithms such as SARSA and Q-learning. We propose a new notion of action value defined by a Gaussian smoothed version of the expected Q-value. We show that such smoothed Q-values still satisfy a Bellman equation, making them learnable from experience sampled from an environment. Moreover, the gradients of expected reward with respect to the mean and covariance of a parameterized Gaussian policy can be recovered from the gradient and Hessian of the smoothed Q-value function. Bas"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1803.02348","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":"1803.02348","created_at":"2026-05-18T00:09:52.557807+00:00"},{"alias_kind":"arxiv_version","alias_value":"1803.02348v3","created_at":"2026-05-18T00:09:52.557807+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1803.02348","created_at":"2026-05-18T00:09:52.557807+00:00"},{"alias_kind":"pith_short_12","alias_value":"NK3FHRXT6CKK","created_at":"2026-05-18T12:32:40.477152+00:00"},{"alias_kind":"pith_short_16","alias_value":"NK3FHRXT6CKKJ7LW","created_at":"2026-05-18T12:32:40.477152+00:00"},{"alias_kind":"pith_short_8","alias_value":"NK3FHRXT","created_at":"2026-05-18T12:32:40.477152+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/NK3FHRXT6CKKJ7LWV3BM4QHEQ2","json":"https://pith.science/pith/NK3FHRXT6CKKJ7LWV3BM4QHEQ2.json","graph_json":"https://pith.science/api/pith-number/NK3FHRXT6CKKJ7LWV3BM4QHEQ2/graph.json","events_json":"https://pith.science/api/pith-number/NK3FHRXT6CKKJ7LWV3BM4QHEQ2/events.json","paper":"https://pith.science/paper/NK3FHRXT"},"agent_actions":{"view_html":"https://pith.science/pith/NK3FHRXT6CKKJ7LWV3BM4QHEQ2","download_json":"https://pith.science/pith/NK3FHRXT6CKKJ7LWV3BM4QHEQ2.json","view_paper":"https://pith.science/paper/NK3FHRXT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1803.02348&json=true","fetch_graph":"https://pith.science/api/pith-number/NK3FHRXT6CKKJ7LWV3BM4QHEQ2/graph.json","fetch_events":"https://pith.science/api/pith-number/NK3FHRXT6CKKJ7LWV3BM4QHEQ2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NK3FHRXT6CKKJ7LWV3BM4QHEQ2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NK3FHRXT6CKKJ7LWV3BM4QHEQ2/action/storage_attestation","attest_author":"https://pith.science/pith/NK3FHRXT6CKKJ7LWV3BM4QHEQ2/action/author_attestation","sign_citation":"https://pith.science/pith/NK3FHRXT6CKKJ7LWV3BM4QHEQ2/action/citation_signature","submit_replication":"https://pith.science/pith/NK3FHRXT6CKKJ7LWV3BM4QHEQ2/action/replication_record"}},"created_at":"2026-05-18T00:09:52.557807+00:00","updated_at":"2026-05-18T00:09:52.557807+00:00"}