{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:Z4CNLNQEJV7DZX3W4RBEGNCCBI","short_pith_number":"pith:Z4CNLNQE","schema_version":"1.0","canonical_sha256":"cf04d5b6044d7e3cdf76e4424334420a151625ff04b80bf3c89e6405897f4175","source":{"kind":"arxiv","id":"2203.04439","version":1},"attestation_state":"computed","paper":{"title":"$\\mathrm{SO}(2)$-Equivariant Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Dian Wang, Robert Platt, Robin Walters","submitted_at":"2022-03-08T23:09:25Z","abstract_excerpt":"Equivariant neural networks enforce symmetry within the structure of their convolutional layers, resulting in a substantial improvement in sample efficiency when learning an equivariant or invariant function. Such models are applicable to robotic manipulation learning which can often be formulated as a rotationally symmetric problem. This paper studies equivariant model architectures in the context of $Q$-learning and actor-critic reinforcement learning. We identify equivariant and invariant characteristics of the optimal $Q$-function and the optimal policy and propose equivariant DQN and SAC "},"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":"2203.04439","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2022-03-08T23:09:25Z","cross_cats_sorted":[],"title_canon_sha256":"0e60c422d267290c15c201d3ade9475fcaf6b6dde96a3e111fd2fe3f0c00da8c","abstract_canon_sha256":"4b4f66e08b541418d0826cb048b4cf8706ab012d110f7efd62891179191dcb54"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:03:18.391919Z","signature_b64":"uX9bxSL0YAm8uykUqHzZcSyxeYlzEB9Nv4LlniB21DwjX82F1jiC5dzB+uDdnK8Tiy4IoM4q8HsUZR4gcjvyCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cf04d5b6044d7e3cdf76e4424334420a151625ff04b80bf3c89e6405897f4175","last_reissued_at":"2026-07-05T04:03:18.391581Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:03:18.391581Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"$\\mathrm{SO}(2)$-Equivariant Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Dian Wang, Robert Platt, Robin Walters","submitted_at":"2022-03-08T23:09:25Z","abstract_excerpt":"Equivariant neural networks enforce symmetry within the structure of their convolutional layers, resulting in a substantial improvement in sample efficiency when learning an equivariant or invariant function. Such models are applicable to robotic manipulation learning which can often be formulated as a rotationally symmetric problem. This paper studies equivariant model architectures in the context of $Q$-learning and actor-critic reinforcement learning. We identify equivariant and invariant characteristics of the optimal $Q$-function and the optimal policy and propose equivariant DQN and SAC "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.04439","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/2203.04439/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":"2203.04439","created_at":"2026-07-05T04:03:18.391630+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.04439v1","created_at":"2026-07-05T04:03:18.391630+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.04439","created_at":"2026-07-05T04:03:18.391630+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z4CNLNQEJV7D","created_at":"2026-07-05T04:03:18.391630+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z4CNLNQEJV7DZX3W","created_at":"2026-07-05T04:03:18.391630+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z4CNLNQE","created_at":"2026-07-05T04:03:18.391630+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24946","citing_title":"Conformal Orbit-Valid Trust Horizons for Equivariant World Models","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2606.19784","citing_title":"EquiVLA: A General Framework for Rotationally Equivariant Vision-Language-Action Models","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Z4CNLNQEJV7DZX3W4RBEGNCCBI","json":"https://pith.science/pith/Z4CNLNQEJV7DZX3W4RBEGNCCBI.json","graph_json":"https://pith.science/api/pith-number/Z4CNLNQEJV7DZX3W4RBEGNCCBI/graph.json","events_json":"https://pith.science/api/pith-number/Z4CNLNQEJV7DZX3W4RBEGNCCBI/events.json","paper":"https://pith.science/paper/Z4CNLNQE"},"agent_actions":{"view_html":"https://pith.science/pith/Z4CNLNQEJV7DZX3W4RBEGNCCBI","download_json":"https://pith.science/pith/Z4CNLNQEJV7DZX3W4RBEGNCCBI.json","view_paper":"https://pith.science/paper/Z4CNLNQE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.04439&json=true","fetch_graph":"https://pith.science/api/pith-number/Z4CNLNQEJV7DZX3W4RBEGNCCBI/graph.json","fetch_events":"https://pith.science/api/pith-number/Z4CNLNQEJV7DZX3W4RBEGNCCBI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z4CNLNQEJV7DZX3W4RBEGNCCBI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z4CNLNQEJV7DZX3W4RBEGNCCBI/action/storage_attestation","attest_author":"https://pith.science/pith/Z4CNLNQEJV7DZX3W4RBEGNCCBI/action/author_attestation","sign_citation":"https://pith.science/pith/Z4CNLNQEJV7DZX3W4RBEGNCCBI/action/citation_signature","submit_replication":"https://pith.science/pith/Z4CNLNQEJV7DZX3W4RBEGNCCBI/action/replication_record"}},"created_at":"2026-07-05T04:03:18.391630+00:00","updated_at":"2026-07-05T04:03:18.391630+00:00"}