{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:IESIFR5XEGOMHBHIVXYSORP3O6","short_pith_number":"pith:IESIFR5X","schema_version":"1.0","canonical_sha256":"412482c7b7219cc384e8adf12745fb778e1290bdb1c8f8f6ba6f791706b5e2ca","source":{"kind":"arxiv","id":"2502.01558","version":2},"attestation_state":"computed","paper":{"title":"Search-Based Adversarial Estimates for Improving Sample Efficiency in Off-Policy Reinforcement Learning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Federico Malato, Ville Hautamaki","submitted_at":"2025-02-03T17:41:02Z","abstract_excerpt":"Sample inefficiency is a long-lasting challenge in deep reinforcement learning (DRL). Despite dramatic improvements have been made, the problem is far from being solved and is especially challenging in environments with sparse or delayed rewards. In our work, we propose to use Adversarial Estimates as a new, simple and efficient approach to mitigate this problem for a class of feedback-based DRL algorithms. Our approach leverages latent similarity search from a small set of human-collected trajectories to boost learning, using only five minutes of human-recorded experience. The results of our "},"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":"2502.01558","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-03T17:41:02Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d43da269ae16cfde7e365c633e38dd9368f20aba8c07b89ffe7d4ed052440b44","abstract_canon_sha256":"e55dd3cf01aa20f3daa3a19b586ca024ca7e38a48583b868e7ed83b75de317d8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:21:14.995955Z","signature_b64":"NNB8bIe8PZBKpOaT9rA/wTs8aHxcmduSo7V6fHTrYJytpFRWY7fPz9iW6lHJrKNg1mKxP3GPE2zGbkfl15f8BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"412482c7b7219cc384e8adf12745fb778e1290bdb1c8f8f6ba6f791706b5e2ca","last_reissued_at":"2026-07-05T11:21:14.995464Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:21:14.995464Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Search-Based Adversarial Estimates for Improving Sample Efficiency in Off-Policy Reinforcement Learning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Federico Malato, Ville Hautamaki","submitted_at":"2025-02-03T17:41:02Z","abstract_excerpt":"Sample inefficiency is a long-lasting challenge in deep reinforcement learning (DRL). Despite dramatic improvements have been made, the problem is far from being solved and is especially challenging in environments with sparse or delayed rewards. In our work, we propose to use Adversarial Estimates as a new, simple and efficient approach to mitigate this problem for a class of feedback-based DRL algorithms. Our approach leverages latent similarity search from a small set of human-collected trajectories to boost learning, using only five minutes of human-recorded experience. The results of our "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.01558","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/2502.01558/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":"2502.01558","created_at":"2026-07-05T11:21:14.995527+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.01558v2","created_at":"2026-07-05T11:21:14.995527+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.01558","created_at":"2026-07-05T11:21:14.995527+00:00"},{"alias_kind":"pith_short_12","alias_value":"IESIFR5XEGOM","created_at":"2026-07-05T11:21:14.995527+00:00"},{"alias_kind":"pith_short_16","alias_value":"IESIFR5XEGOMHBHI","created_at":"2026-07-05T11:21:14.995527+00:00"},{"alias_kind":"pith_short_8","alias_value":"IESIFR5X","created_at":"2026-07-05T11:21:14.995527+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.05529","citing_title":"Avoiding Death through Fear Intrinsic Conditioning","ref_index":20,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IESIFR5XEGOMHBHIVXYSORP3O6","json":"https://pith.science/pith/IESIFR5XEGOMHBHIVXYSORP3O6.json","graph_json":"https://pith.science/api/pith-number/IESIFR5XEGOMHBHIVXYSORP3O6/graph.json","events_json":"https://pith.science/api/pith-number/IESIFR5XEGOMHBHIVXYSORP3O6/events.json","paper":"https://pith.science/paper/IESIFR5X"},"agent_actions":{"view_html":"https://pith.science/pith/IESIFR5XEGOMHBHIVXYSORP3O6","download_json":"https://pith.science/pith/IESIFR5XEGOMHBHIVXYSORP3O6.json","view_paper":"https://pith.science/paper/IESIFR5X","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.01558&json=true","fetch_graph":"https://pith.science/api/pith-number/IESIFR5XEGOMHBHIVXYSORP3O6/graph.json","fetch_events":"https://pith.science/api/pith-number/IESIFR5XEGOMHBHIVXYSORP3O6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IESIFR5XEGOMHBHIVXYSORP3O6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IESIFR5XEGOMHBHIVXYSORP3O6/action/storage_attestation","attest_author":"https://pith.science/pith/IESIFR5XEGOMHBHIVXYSORP3O6/action/author_attestation","sign_citation":"https://pith.science/pith/IESIFR5XEGOMHBHIVXYSORP3O6/action/citation_signature","submit_replication":"https://pith.science/pith/IESIFR5XEGOMHBHIVXYSORP3O6/action/replication_record"}},"created_at":"2026-07-05T11:21:14.995527+00:00","updated_at":"2026-07-05T11:21:14.995527+00:00"}