{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:GDSQ2VPGEOCNC635W3KME3QWUP","short_pith_number":"pith:GDSQ2VPG","schema_version":"1.0","canonical_sha256":"30e50d55e62384d17b7db6d4c26e16a3c7a80b663d02bc223f52fd3fea990922","source":{"kind":"arxiv","id":"2103.06386","version":1},"attestation_state":"computed","paper":{"title":"Improving Context-Based Meta-Reinforcement Learning with Self-Supervised Trajectory Contrastive Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bernie Wang, Bichen Wu, Kurt Keutzer, Simon Xu, Yang Gao","submitted_at":"2021-03-10T23:31:19Z","abstract_excerpt":"Meta-reinforcement learning typically requires orders of magnitude more samples than single task reinforcement learning methods. This is because meta-training needs to deal with more diverse distributions and train extra components such as context encoders. To address this, we propose a novel self-supervised learning task, which we named Trajectory Contrastive Learning (TCL), to improve meta-training. TCL adopts contrastive learning and trains a context encoder to predict whether two transition windows are sampled from the same trajectory. TCL leverages the natural hierarchical structure of co"},"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":"2103.06386","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-03-10T23:31:19Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ebdbf213dcd21486914506e492f8cd166f650b61f45b16649e83ac2f49612b01","abstract_canon_sha256":"e01bd97fd686350223139365dbc8ff8297eb0ff1d440ca689da9821a9ef8b596"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:22:08.605449Z","signature_b64":"oH0mZhf4HTUbJB45+lpVPGWL9JIwFTOcMmWfK1ZH5czU6GnkalcuOY3JVm1Q48fWgRcvTrf8mG8DRSYCUbrZAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"30e50d55e62384d17b7db6d4c26e16a3c7a80b663d02bc223f52fd3fea990922","last_reissued_at":"2026-07-05T02:22:08.604979Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:22:08.604979Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Context-Based Meta-Reinforcement Learning with Self-Supervised Trajectory Contrastive Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bernie Wang, Bichen Wu, Kurt Keutzer, Simon Xu, Yang Gao","submitted_at":"2021-03-10T23:31:19Z","abstract_excerpt":"Meta-reinforcement learning typically requires orders of magnitude more samples than single task reinforcement learning methods. This is because meta-training needs to deal with more diverse distributions and train extra components such as context encoders. To address this, we propose a novel self-supervised learning task, which we named Trajectory Contrastive Learning (TCL), to improve meta-training. TCL adopts contrastive learning and trains a context encoder to predict whether two transition windows are sampled from the same trajectory. TCL leverages the natural hierarchical structure of co"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.06386","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/2103.06386/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":"2103.06386","created_at":"2026-07-05T02:22:08.605035+00:00"},{"alias_kind":"arxiv_version","alias_value":"2103.06386v1","created_at":"2026-07-05T02:22:08.605035+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.06386","created_at":"2026-07-05T02:22:08.605035+00:00"},{"alias_kind":"pith_short_12","alias_value":"GDSQ2VPGEOCN","created_at":"2026-07-05T02:22:08.605035+00:00"},{"alias_kind":"pith_short_16","alias_value":"GDSQ2VPGEOCNC635","created_at":"2026-07-05T02:22:08.605035+00:00"},{"alias_kind":"pith_short_8","alias_value":"GDSQ2VPG","created_at":"2026-07-05T02:22:08.605035+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/GDSQ2VPGEOCNC635W3KME3QWUP","json":"https://pith.science/pith/GDSQ2VPGEOCNC635W3KME3QWUP.json","graph_json":"https://pith.science/api/pith-number/GDSQ2VPGEOCNC635W3KME3QWUP/graph.json","events_json":"https://pith.science/api/pith-number/GDSQ2VPGEOCNC635W3KME3QWUP/events.json","paper":"https://pith.science/paper/GDSQ2VPG"},"agent_actions":{"view_html":"https://pith.science/pith/GDSQ2VPGEOCNC635W3KME3QWUP","download_json":"https://pith.science/pith/GDSQ2VPGEOCNC635W3KME3QWUP.json","view_paper":"https://pith.science/paper/GDSQ2VPG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2103.06386&json=true","fetch_graph":"https://pith.science/api/pith-number/GDSQ2VPGEOCNC635W3KME3QWUP/graph.json","fetch_events":"https://pith.science/api/pith-number/GDSQ2VPGEOCNC635W3KME3QWUP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GDSQ2VPGEOCNC635W3KME3QWUP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GDSQ2VPGEOCNC635W3KME3QWUP/action/storage_attestation","attest_author":"https://pith.science/pith/GDSQ2VPGEOCNC635W3KME3QWUP/action/author_attestation","sign_citation":"https://pith.science/pith/GDSQ2VPGEOCNC635W3KME3QWUP/action/citation_signature","submit_replication":"https://pith.science/pith/GDSQ2VPGEOCNC635W3KME3QWUP/action/replication_record"}},"created_at":"2026-07-05T02:22:08.605035+00:00","updated_at":"2026-07-05T02:22:08.605035+00:00"}