{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:BUNORPAJ3B6XJTI6X4DTJEE6ET","short_pith_number":"pith:BUNORPAJ","schema_version":"1.0","canonical_sha256":"0d1ae8bc09d87d74cd1ebf0734909e24c05c54347b71bab2fcca4156702f87b7","source":{"kind":"arxiv","id":"2002.08550","version":3},"attestation_state":"computed","paper":{"title":"Learning to Walk in the Real World with Minimal Human Effort","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"Jie Tan, Peng Xu, Sehoon Ha, Sergey Levine, Zhenyu Tan","submitted_at":"2020-02-20T03:36:39Z","abstract_excerpt":"Reliable and stable locomotion has been one of the most fundamental challenges for legged robots. Deep reinforcement learning (deep RL) has emerged as a promising method for developing such control policies autonomously. In this paper, we develop a system for learning legged locomotion policies with deep RL in the real world with minimal human effort. The key difficulties for on-robot learning systems are automatic data collection and safety. We overcome these two challenges by developing a multi-task learning procedure and a safety-constrained RL framework. We tested our system on the task of"},"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":"2002.08550","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2020-02-20T03:36:39Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"441350bfde98ed00e6d74cb43d4cd3cf300a8b44266466d5c9ea27fc1318e051","abstract_canon_sha256":"1d97650cde324aa09dc27b242abefc73bba7e597df9b2ce31e73712928ed1cbb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:48:25.894786Z","signature_b64":"RWpUQyT9ltV2FH/BXh+zZ6cgexADT9XPcAHvNtqIyJhoXt9zqdNYTYAKACKO5P2b0lek77YjeqzFVdavFaviDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0d1ae8bc09d87d74cd1ebf0734909e24c05c54347b71bab2fcca4156702f87b7","last_reissued_at":"2026-07-05T01:48:25.894349Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:48:25.894349Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning to Walk in the Real World with Minimal Human Effort","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"Jie Tan, Peng Xu, Sehoon Ha, Sergey Levine, Zhenyu Tan","submitted_at":"2020-02-20T03:36:39Z","abstract_excerpt":"Reliable and stable locomotion has been one of the most fundamental challenges for legged robots. Deep reinforcement learning (deep RL) has emerged as a promising method for developing such control policies autonomously. In this paper, we develop a system for learning legged locomotion policies with deep RL in the real world with minimal human effort. The key difficulties for on-robot learning systems are automatic data collection and safety. We overcome these two challenges by developing a multi-task learning procedure and a safety-constrained RL framework. We tested our system on the task of"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2002.08550","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2002.08550/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":"2002.08550","created_at":"2026-07-05T01:48:25.894409+00:00"},{"alias_kind":"arxiv_version","alias_value":"2002.08550v3","created_at":"2026-07-05T01:48:25.894409+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2002.08550","created_at":"2026-07-05T01:48:25.894409+00:00"},{"alias_kind":"pith_short_12","alias_value":"BUNORPAJ3B6X","created_at":"2026-07-05T01:48:25.894409+00:00"},{"alias_kind":"pith_short_16","alias_value":"BUNORPAJ3B6XJTI6","created_at":"2026-07-05T01:48:25.894409+00:00"},{"alias_kind":"pith_short_8","alias_value":"BUNORPAJ","created_at":"2026-07-05T01:48:25.894409+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.09595","citing_title":"Neuromorphic Reinforcement Learning for Quadruped Locomotion Control on Uneven Terrain","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2606.31320","citing_title":"Safe Online Learning via Smooth Safety-Structured Policy Composition","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09595","citing_title":"Neuromorphic Reinforcement Learning for Quadruped Locomotion Control on Uneven Terrain","ref_index":8,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BUNORPAJ3B6XJTI6X4DTJEE6ET","json":"https://pith.science/pith/BUNORPAJ3B6XJTI6X4DTJEE6ET.json","graph_json":"https://pith.science/api/pith-number/BUNORPAJ3B6XJTI6X4DTJEE6ET/graph.json","events_json":"https://pith.science/api/pith-number/BUNORPAJ3B6XJTI6X4DTJEE6ET/events.json","paper":"https://pith.science/paper/BUNORPAJ"},"agent_actions":{"view_html":"https://pith.science/pith/BUNORPAJ3B6XJTI6X4DTJEE6ET","download_json":"https://pith.science/pith/BUNORPAJ3B6XJTI6X4DTJEE6ET.json","view_paper":"https://pith.science/paper/BUNORPAJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2002.08550&json=true","fetch_graph":"https://pith.science/api/pith-number/BUNORPAJ3B6XJTI6X4DTJEE6ET/graph.json","fetch_events":"https://pith.science/api/pith-number/BUNORPAJ3B6XJTI6X4DTJEE6ET/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BUNORPAJ3B6XJTI6X4DTJEE6ET/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BUNORPAJ3B6XJTI6X4DTJEE6ET/action/storage_attestation","attest_author":"https://pith.science/pith/BUNORPAJ3B6XJTI6X4DTJEE6ET/action/author_attestation","sign_citation":"https://pith.science/pith/BUNORPAJ3B6XJTI6X4DTJEE6ET/action/citation_signature","submit_replication":"https://pith.science/pith/BUNORPAJ3B6XJTI6X4DTJEE6ET/action/replication_record"}},"created_at":"2026-07-05T01:48:25.894409+00:00","updated_at":"2026-07-05T01:48:25.894409+00:00"}