{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:DV2GB2SDS6TGAENFDUSUI4ZUCY","short_pith_number":"pith:DV2GB2SD","schema_version":"1.0","canonical_sha256":"1d7460ea4397a66011a51d2544733416360a8cedf0e6cd94e9a78841dc715288","source":{"kind":"arxiv","id":"1702.08165","version":2},"attestation_state":"computed","paper":{"title":"Reinforcement Learning with Deep Energy-Based Policies","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Haoran Tang, Pieter Abbeel, Sergey Levine, Tuomas Haarnoja","submitted_at":"2017-02-27T07:16:41Z","abstract_excerpt":"We propose a method for learning expressive energy-based policies for continuous states and actions, which has been feasible only in tabular domains before. We apply our method to learning maximum entropy policies, resulting into a new algorithm, called soft Q-learning, that expresses the optimal policy via a Boltzmann distribution. We use the recently proposed amortized Stein variational gradient descent to learn a stochastic sampling network that approximates samples from this distribution. The benefits of the proposed algorithm include improved exploration and compositionality that allows t"},"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":"1702.08165","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-02-27T07:16:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a56bd6d19392a5a576d600fe2dd87fdd8d70fb957a236a72781171e183a2564d","abstract_canon_sha256":"850219589e3c015b7b5a9359f19f36275fe231e3eab2a577757ddaa05530436e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:39:46.786915Z","signature_b64":"eEnRqDn+f4ax2A0pFsYwKSWPUOqXW8ZpPUCWmgVvhkFCf+XYki550iMaGGxY+yV3gVjS3o4J9rIVJxiCEcZaCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1d7460ea4397a66011a51d2544733416360a8cedf0e6cd94e9a78841dc715288","last_reissued_at":"2026-05-18T00:39:46.786245Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:39:46.786245Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reinforcement Learning with Deep Energy-Based Policies","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Haoran Tang, Pieter Abbeel, Sergey Levine, Tuomas Haarnoja","submitted_at":"2017-02-27T07:16:41Z","abstract_excerpt":"We propose a method for learning expressive energy-based policies for continuous states and actions, which has been feasible only in tabular domains before. We apply our method to learning maximum entropy policies, resulting into a new algorithm, called soft Q-learning, that expresses the optimal policy via a Boltzmann distribution. We use the recently proposed amortized Stein variational gradient descent to learn a stochastic sampling network that approximates samples from this distribution. The benefits of the proposed algorithm include improved exploration and compositionality that allows t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1702.08165","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":""},"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":"1702.08165","created_at":"2026-05-18T00:39:46.786364+00:00"},{"alias_kind":"arxiv_version","alias_value":"1702.08165v2","created_at":"2026-05-18T00:39:46.786364+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1702.08165","created_at":"2026-05-18T00:39:46.786364+00:00"},{"alias_kind":"pith_short_12","alias_value":"DV2GB2SDS6TG","created_at":"2026-05-18T12:31:12.930513+00:00"},{"alias_kind":"pith_short_16","alias_value":"DV2GB2SDS6TGAENF","created_at":"2026-05-18T12:31:12.930513+00:00"},{"alias_kind":"pith_short_8","alias_value":"DV2GB2SD","created_at":"2026-05-18T12:31:12.930513+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"2605.23565","citing_title":"Understanding Goal Generalisation in Sequential Reinforcement Learning","ref_index":25,"is_internal_anchor":true},{"citing_arxiv_id":"1907.06143","citing_title":"Neural Embedding for Physical Manipulations","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2509.25424","citing_title":"Polychromic Objectives for Reinforcement Learning","ref_index":10,"is_internal_anchor":true},{"citing_arxiv_id":"2604.19695","citing_title":"Planning in entropy-regularized Markov decision processes and games","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DV2GB2SDS6TGAENFDUSUI4ZUCY","json":"https://pith.science/pith/DV2GB2SDS6TGAENFDUSUI4ZUCY.json","graph_json":"https://pith.science/api/pith-number/DV2GB2SDS6TGAENFDUSUI4ZUCY/graph.json","events_json":"https://pith.science/api/pith-number/DV2GB2SDS6TGAENFDUSUI4ZUCY/events.json","paper":"https://pith.science/paper/DV2GB2SD"},"agent_actions":{"view_html":"https://pith.science/pith/DV2GB2SDS6TGAENFDUSUI4ZUCY","download_json":"https://pith.science/pith/DV2GB2SDS6TGAENFDUSUI4ZUCY.json","view_paper":"https://pith.science/paper/DV2GB2SD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1702.08165&json=true","fetch_graph":"https://pith.science/api/pith-number/DV2GB2SDS6TGAENFDUSUI4ZUCY/graph.json","fetch_events":"https://pith.science/api/pith-number/DV2GB2SDS6TGAENFDUSUI4ZUCY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DV2GB2SDS6TGAENFDUSUI4ZUCY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DV2GB2SDS6TGAENFDUSUI4ZUCY/action/storage_attestation","attest_author":"https://pith.science/pith/DV2GB2SDS6TGAENFDUSUI4ZUCY/action/author_attestation","sign_citation":"https://pith.science/pith/DV2GB2SDS6TGAENFDUSUI4ZUCY/action/citation_signature","submit_replication":"https://pith.science/pith/DV2GB2SDS6TGAENFDUSUI4ZUCY/action/replication_record"}},"created_at":"2026-05-18T00:39:46.786364+00:00","updated_at":"2026-05-18T00:39:46.786364+00:00"}