{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:M7Y6GROMDKPL35XWLJYHRL6LST","short_pith_number":"pith:M7Y6GROM","schema_version":"1.0","canonical_sha256":"67f1e345cc1a9ebdf6f65a7078afcb94e98123f872a5912132c4f1fa853e64a4","source":{"kind":"arxiv","id":"2503.10318","version":1},"attestation_state":"computed","paper":{"title":"Enhance Exploration in Safe Reinforcement Learning with Contrastive Representation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bang Giang Le, Duc Kien Doan, Viet Cuong Ta","submitted_at":"2025-03-13T12:53:42Z","abstract_excerpt":"In safe reinforcement learning, agent needs to balance between exploration actions and safety constraints. Following this paradigm, domain transfer approaches learn a prior Q-function from the related environments to prevent unsafe actions. However, because of the large number of false positives, some safe actions are never executed, leading to inadequate exploration in sparse-reward environments. In this work, we aim to learn an efficient state representation to balance the exploration and safety-prefer action in a sparse-reward environment. Firstly, the image input is mapped to latent repres"},"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":"2503.10318","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-03-13T12:53:42Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"e01bc8ae832a9295e34413d37efc3f58d9442c4b45a6eb06e956691304fe0f14","abstract_canon_sha256":"db388e514a70ce41e14d376710fbda62e2077619509cec14a027a67bdda5097c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:54:15.600917Z","signature_b64":"BFQW94ZPD1ZTanVvZKLmyJtRnkunpWaxWsA2Rf+SA6SUajsNVwJvVSCKnYL+RMM0KyouxVn+SZivoR8960NRBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"67f1e345cc1a9ebdf6f65a7078afcb94e98123f872a5912132c4f1fa853e64a4","last_reissued_at":"2026-07-05T10:54:15.600420Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:54:15.600420Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Enhance Exploration in Safe Reinforcement Learning with Contrastive Representation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bang Giang Le, Duc Kien Doan, Viet Cuong Ta","submitted_at":"2025-03-13T12:53:42Z","abstract_excerpt":"In safe reinforcement learning, agent needs to balance between exploration actions and safety constraints. Following this paradigm, domain transfer approaches learn a prior Q-function from the related environments to prevent unsafe actions. However, because of the large number of false positives, some safe actions are never executed, leading to inadequate exploration in sparse-reward environments. In this work, we aim to learn an efficient state representation to balance the exploration and safety-prefer action in a sparse-reward environment. Firstly, the image input is mapped to latent repres"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.10318","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/2503.10318/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":"2503.10318","created_at":"2026-07-05T10:54:15.600480+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.10318v1","created_at":"2026-07-05T10:54:15.600480+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.10318","created_at":"2026-07-05T10:54:15.600480+00:00"},{"alias_kind":"pith_short_12","alias_value":"M7Y6GROMDKPL","created_at":"2026-07-05T10:54:15.600480+00:00"},{"alias_kind":"pith_short_16","alias_value":"M7Y6GROMDKPL35XW","created_at":"2026-07-05T10:54:15.600480+00:00"},{"alias_kind":"pith_short_8","alias_value":"M7Y6GROM","created_at":"2026-07-05T10:54:15.600480+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/M7Y6GROMDKPL35XWLJYHRL6LST","json":"https://pith.science/pith/M7Y6GROMDKPL35XWLJYHRL6LST.json","graph_json":"https://pith.science/api/pith-number/M7Y6GROMDKPL35XWLJYHRL6LST/graph.json","events_json":"https://pith.science/api/pith-number/M7Y6GROMDKPL35XWLJYHRL6LST/events.json","paper":"https://pith.science/paper/M7Y6GROM"},"agent_actions":{"view_html":"https://pith.science/pith/M7Y6GROMDKPL35XWLJYHRL6LST","download_json":"https://pith.science/pith/M7Y6GROMDKPL35XWLJYHRL6LST.json","view_paper":"https://pith.science/paper/M7Y6GROM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.10318&json=true","fetch_graph":"https://pith.science/api/pith-number/M7Y6GROMDKPL35XWLJYHRL6LST/graph.json","fetch_events":"https://pith.science/api/pith-number/M7Y6GROMDKPL35XWLJYHRL6LST/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/M7Y6GROMDKPL35XWLJYHRL6LST/action/timestamp_anchor","attest_storage":"https://pith.science/pith/M7Y6GROMDKPL35XWLJYHRL6LST/action/storage_attestation","attest_author":"https://pith.science/pith/M7Y6GROMDKPL35XWLJYHRL6LST/action/author_attestation","sign_citation":"https://pith.science/pith/M7Y6GROMDKPL35XWLJYHRL6LST/action/citation_signature","submit_replication":"https://pith.science/pith/M7Y6GROMDKPL35XWLJYHRL6LST/action/replication_record"}},"created_at":"2026-07-05T10:54:15.600480+00:00","updated_at":"2026-07-05T10:54:15.600480+00:00"}