{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:TCMMLHA7FOZ2QXXRZ4TFZ77V5Q","short_pith_number":"pith:TCMMLHA7","schema_version":"1.0","canonical_sha256":"9898c59c1f2bb3a85ef1cf265cfff5ec14fa9508fa62e9a6829f24c896cbbaec","source":{"kind":"arxiv","id":"2502.02316","version":2},"attestation_state":"computed","paper":{"title":"DIME:Diffusion-Based Maximum Entropy Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Daniel Palenicek, Denis Blessing, Ge Li, Georgia Chalvatzaki, Gerhard Neumann, Jan Peters, Onur Celik, Zechu Li","submitted_at":"2025-02-04T13:37:14Z","abstract_excerpt":"Maximum entropy reinforcement learning (MaxEnt-RL) has become the standard approach to RL due to its beneficial exploration properties. Traditionally, policies are parameterized using Gaussian distributions, which significantly limits their representational capacity. Diffusion-based policies offer a more expressive alternative, yet integrating them into MaxEnt-RL poses challenges-primarily due to the intractability of computing their marginal entropy. To overcome this, we propose Diffusion-Based Maximum Entropy RL (DIME). \\emph{DIME} leverages recent advances in approximate inference with diff"},"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":"2502.02316","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-04T13:37:14Z","cross_cats_sorted":[],"title_canon_sha256":"a21b9a2b8edc22978e097b4852bc5012c0ebb2c4fa89bfa92c65a49559e296f6","abstract_canon_sha256":"2cdd74a8144e7b034df2dd0bdd50181693d4c55e54ddecd19f6a19f514e7e5fa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:19:03.270282Z","signature_b64":"vjsvglqua2XRYPO/2af7b579iqcKUWK087QfegQs0fl1kM8u9IbKMi6hcpn8vTxKK3tPjzLPxyJ7g8JVGNCICA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9898c59c1f2bb3a85ef1cf265cfff5ec14fa9508fa62e9a6829f24c896cbbaec","last_reissued_at":"2026-07-05T11:19:03.269813Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:19:03.269813Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DIME:Diffusion-Based Maximum Entropy Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Daniel Palenicek, Denis Blessing, Ge Li, Georgia Chalvatzaki, Gerhard Neumann, Jan Peters, Onur Celik, Zechu Li","submitted_at":"2025-02-04T13:37:14Z","abstract_excerpt":"Maximum entropy reinforcement learning (MaxEnt-RL) has become the standard approach to RL due to its beneficial exploration properties. Traditionally, policies are parameterized using Gaussian distributions, which significantly limits their representational capacity. Diffusion-based policies offer a more expressive alternative, yet integrating them into MaxEnt-RL poses challenges-primarily due to the intractability of computing their marginal entropy. To overcome this, we propose Diffusion-Based Maximum Entropy RL (DIME). \\emph{DIME} leverages recent advances in approximate inference with diff"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.02316","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/2502.02316/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":"2502.02316","created_at":"2026-07-05T11:19:03.269875+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.02316v2","created_at":"2026-07-05T11:19:03.269875+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.02316","created_at":"2026-07-05T11:19:03.269875+00:00"},{"alias_kind":"pith_short_12","alias_value":"TCMMLHA7FOZ2","created_at":"2026-07-05T11:19:03.269875+00:00"},{"alias_kind":"pith_short_16","alias_value":"TCMMLHA7FOZ2QXXR","created_at":"2026-07-05T11:19:03.269875+00:00"},{"alias_kind":"pith_short_8","alias_value":"TCMMLHA7","created_at":"2026-07-05T11:19:03.269875+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22630","citing_title":"Scalable Maximum Entropy Reinforcement Learning for Diffusion Policies via Adjoint Matching","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06967","citing_title":"GenPO++: Generative Policy Optimization with Jacobian-free Likelihood Ratios","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2509.22963","citing_title":"Reinforcement Learning with Discrete Diffusion Policies for Combinatorial Action Spaces","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2603.04333","citing_title":"What Does Flow Matching Bring To TD Learning?","ref_index":8,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TCMMLHA7FOZ2QXXRZ4TFZ77V5Q","json":"https://pith.science/pith/TCMMLHA7FOZ2QXXRZ4TFZ77V5Q.json","graph_json":"https://pith.science/api/pith-number/TCMMLHA7FOZ2QXXRZ4TFZ77V5Q/graph.json","events_json":"https://pith.science/api/pith-number/TCMMLHA7FOZ2QXXRZ4TFZ77V5Q/events.json","paper":"https://pith.science/paper/TCMMLHA7"},"agent_actions":{"view_html":"https://pith.science/pith/TCMMLHA7FOZ2QXXRZ4TFZ77V5Q","download_json":"https://pith.science/pith/TCMMLHA7FOZ2QXXRZ4TFZ77V5Q.json","view_paper":"https://pith.science/paper/TCMMLHA7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.02316&json=true","fetch_graph":"https://pith.science/api/pith-number/TCMMLHA7FOZ2QXXRZ4TFZ77V5Q/graph.json","fetch_events":"https://pith.science/api/pith-number/TCMMLHA7FOZ2QXXRZ4TFZ77V5Q/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TCMMLHA7FOZ2QXXRZ4TFZ77V5Q/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TCMMLHA7FOZ2QXXRZ4TFZ77V5Q/action/storage_attestation","attest_author":"https://pith.science/pith/TCMMLHA7FOZ2QXXRZ4TFZ77V5Q/action/author_attestation","sign_citation":"https://pith.science/pith/TCMMLHA7FOZ2QXXRZ4TFZ77V5Q/action/citation_signature","submit_replication":"https://pith.science/pith/TCMMLHA7FOZ2QXXRZ4TFZ77V5Q/action/replication_record"}},"created_at":"2026-07-05T11:19:03.269875+00:00","updated_at":"2026-07-05T11:19:03.269875+00:00"}